Railroad break and train sensor system

Abstract

A railroad track sensor system has: a plurality of sensor units distributed at various respective locations along a section of railroad track that is bounded on opposed ends by rail-to-rail shunts, the plurality of sensor units configured to cycle through a rotation where each sensor unit is set to operate, one at a time, in a sending mode where the sensor unit produces a stimulus electrical voltage across the parallel rails, while the plurality of sensor units are set to simultaneously operate in a receiving mode where the plurality of sensor units each measure a resultant voltage across the parallel rails of the railroad track; and a control unit configured to compare voltage waveform data, from the resultant voltages measured by the plurality of sensor units during the rotation, with calibrated voltage waveform data to determine whether the section of the railroad is in a normal state, a broken rail state, or a train present state.

Description

RAILROAD BREAK AND TRAIN SENSOR SYSTEMTECHNICAL FIELD

[0001] The present document relates railroad break and train sensor systems, and related methods of use. BACKGROUND

[0002] The following paragraphs are not an admission that anything discussed in them is prior art or part of the knowledge of persons skilled in the art.

[0003] Wireless mesh networks have been used for railway video surveillance, on-board systems communication, and troubleshooting. Remote areas pose a challenge for traditional networks to avoid the perils and uncertainties of gaps in cellular or other long-range communication networks.SUMMARY

[0004] A railroad track sensor system is disclosed comprising: a plurality of sensor units distributed at various respective locations along a section of railroad track that is bounded on opposed ends by rail-to-rail shunts, the plurality of sensor units configured to cycle through a rotation where each sensor unit is set to operate, one at a time, in a sending mode where the sensor unit produces a stimulus electrical voltage across the parallel rails, while the plurality of sensor units are set to simultaneously operate in a receiving mode where the plurality of sensor units each measure a resultant voltage across the parallel rails of the railroad track; and a control unit configured to compare voltage waveform data, from the resultant voltages measured by the plurality of sensor units during the rotation, with calibrated voltage waveform data to determine whether the section of the railroad is in a normal state, a broken rail state, or a train present state.

[0005] A method is disclosed of monitoring, and in some cases assessing a status of, a section of a railroad track that is bounded on opposed ends by rail-to-rail shunts, with a plurality of sensor units distributed at various respective locations along the section of railroad track, the method comprising: cycling the plurality of sensor units through a rotation where each sensor unit operates, one at a time, in a sending mode where the sensor unit produces a stimulus electrical voltage across the parallel rails, while the plurality of sensor units are simultaneously operate in a receiving mode where the plurality of sensor units each measure a resultant voltage across the parallel rails of the railroad track; and determining whether the section of the railroad is in a normal state, a broken rail state, or a train present state, by comparing voltage waveform data, from the resultant voltages measured by the plurality of sensor units during the rotation, with calibrated voltage waveform data.

[0006] The disclosed Rail Sensor is used primarily to detect the presence of a break in the rail as indicated by a loss of continuity in a closed loop of track. The system comprises either a single sensor (for short blocks of track) or a plurality of sensors located at intervals ranging from a half a mile to 2 or more miles. The boundaries of the sensing region are defined by shunts across the rails. The shunts can be either normally occurring shunts such as unisolated switches, or specifically placed boundary shunts. The system may provide a stimulus voltage across the adjacent rails and measures the resultant voltage waveform created by the impedance of the track at the Sending unit. If more than one unit is located on the sensing block, the units will each take a turn being the Sending unit while the other units measure the rail-to-rail voltage waveform at each of their individual locations. The units will rotate through the Sending role once per update interval. (See Figure 1). The waveforms of the Sending unit and anyMeasuring unit(s) are sent to a server. The server monitors for changes in the Sending waveform and each of the related waveforms measured at the fixed intervals along the track. When more than one unit is present in a sensing block, the data from the full Role rotation can be aggregated to provide redundancy and improved detection results. The primary conclusions that can be inferred from the data are Normal (Figure 1), Rail Break Detected (Figures 3 and 5) and Train Present (Figure 7). The location of the rail break can be detected as being; between any 2 units on the sensing block; outside of a range defined by any 1 or 2 units on the sending block; or, near a unit on the sensing block. Other diagnostic data can be collected as well. Some examples of diagnostic data include, but not limited to, whether the leads are connected to the rails, shorted leads, interfering electrical noise. A communications system is disclosed comprising: a plurality of nodes distributed within a geographical area, each of the plurality of nodes being in communication with one or more of the other of the plurality of nodes to collectively form at least a partially- connected wireless mesh network; and each of the plurality of nodes having: a node processor; a battery; a connection to a power source; a short-range transceiver for short-range communications between nodes of the plurality of nodes; and a long-range transceiver for long-range communication with a long-range network; and in which each of the plurality of nodes are configured to dynamically assign and use one or more of the plurality of nodes as a border router.

[0007] A method is disclosed comprising: using a plurality of nodes to communicate with each other in a partially connected mesh network and dynamically assign one or more of the plurality of nodes as a border router, in which the plurality of nodes are distributed within a geographical area; and using the border router to relay communications from the plurality of nodes through the long-range network.

[0008] A communications system, according to the present disclosure, has a plurality of trackside nodes, arranged alongside a railway line, each in communication with at least one other of the plurality of nodes. A plurality of gateway routers installed periodically adjacent to the railway line, to facilitate communication between one or more nodes and a long-range network.

[0009] In another embodiment, each node has a first wireless transceiver for short distance communication and a second wireless transceiver for long distance communication. The system software gathers information on performance factors to evaluate each node and determine which nodes will communicate both with other nearby nodes and with the long-range network, in place of the plurality of gateway routers.

[0010] In another embodiment, the performance factors are one or more factors selected from the group consisting of: battery performance, number of other nodes in communication range, cellular signal strength, number of other nearby candidate uplink nodes, and distance to the nearest suitable candidate uplink node.

[0011] In another embodiment, the first and / or second transceiver is a low frequency radio transceiver, for example an IEEE ™ 802.15.4E 2.4 GHz wireless transceiver, a 2GFSK transceiver or a device that operates using other protocols.

[0012] A DC current-based sensor system may monitor changes in voltage in order to assess rail integrity and provide information on the status of a rail line. Sensor units may be provided in groups, with a suitable number of sensor units per group, such as between 1 and 10 units. Sensor units may have the ability to apply relatively low voltage signals to the rail. The sensor units may also measure the voltage applied. Groups are separated using shuntson each end of the monitored track section. Shunts are installed at suitable locations and spacings from sensor units, for example approximately 1 mile from the closest sensor unit within a group. Multiple sensor units within a group may provide overlapping response zones and redundancy. Redundancy may be beneficial as neighbouring sensor units may be used for information on the track segment, including broken rail detection, even in the event that one or more sensor units become non-responsive or non-reporting. Data from a sensor unit that indicates a rail event, such as a rail break, may be corroborated or validated with data from neighboring sensor units. A suitable back-office server may be used or otherwise connected to receive data from the sensor units, for example as part of a cloud system. Data, such as measured voltages may be compared via a suitable algorithm, such as a rolling average algorithm. Alerts, alarms, or other indicator steps may be triggered upon detection of a rail event, such as if there is detected a significant change from the rolling averages for a track segment. A monitoring cycle may be repeated among a group of sensor units throughout a sector of dark territory on the track.

[0013] Systems and methods of reducing alternating current (AC) interference from voltage waveform data measured by the plurality of sensor units are disclosed.

[0014] In various embodiments, there may be included any one or more of the following features: The control unit is configured to determine an event location, being a length of the section of railroad track between adjacent sensor units or between one of the opposed ends of the section of railroad track and an adjacent sensor unit, where a broken rail event or a train present event has occurred. The control unit is configured to determine an event location where a broken rail event has occurred, in use, between sensor units by detecting: elevated voltage waveform data at the sensor units on a first side of the event location, and attenuated voltage waveform data at the sensor units on a second side of the event location, when the sensor units on the first side are in the sending mode; and attenuated voltage waveform data at the sensor units on the first side of the event location, and elevated voltage waveform data at the sensor units on the second side of the event location, when the sensor units on the second side are in the sending mode. The control unit is configured to determine an event location where a broken rail event has occurred, in use, between one of the opposed ends of the section of railroad track and an adjacent sensor unit, by detecting elevated voltage waveform data of the plurality of sensor units, with a relative elevation of the voltage waveform data decreasing, in send mode and receive mode, with increasing distance from the adjacent sensor unit. The control unit is configured to determine that the broken rail event corresponds to a loose joint event by detecting, after initial detection of the broken rail event, relatively less to no, attenuation if detectable, and elevation, of voltage waveform data. The control unit is configured to determine that the broken rail event corresponds to the loose joint event by detecting, after initial detection of the broken rail event, relatively less to no, attenuation if detectable, and elevation, voltage waveform data during neutral zone temperature transitions, such as during daylight hours. The control unit is configured to determine an event location where a train present event has occurred, in use, at an adjacent sensor unit or between adjacent sensor units by detecting attenuated voltage waveform data of the plurality of sensor units in send mode and receive mode. The control unit is configured to determine an event location where a train present event has occurred, in use, between adjacent sensor units by detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from the adjacent sensor units in send mode and receive mode. The control unit is configured to determine an event location where a train present event has occurred, in use, at an adjacent sensor unit by: detecting a relative local maximumattenuation of voltage waveform data at the adjacent sensor unit in send mode; and detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from sensor units adjacent to the adjacent sensor unit. Determining comprises determining that the broken rail event corresponds to a loose joint event by detecting, after initial detection of the broken rail event, relatively less to no, attenuation if detectable, and elevation, of voltage waveform data. The control unit is configmed to determine an event location, being a length of the section of railroad track between two or more sensor units, of the plurality of sensor units, or between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, where one or more of a broken rail event, a train present event, or a loose joint event has occurred. The control unit is configured to determine an event location where a broken rail event has occurred, in use, between sensor units, of the plurality of sensor units, by detecting: elevated voltage waveform data at one or more sensor units, of the plurality of sensor units, on a first side of the event location, and attenuated voltage waveform data at one or more sensor units, of the plurality of sensor units, on a second side of the event location, when the one or more sensor units on the first side are in the sending mode; and attenuated voltage waveform data at the one or more sensor units on the first side of the event location, and elevated voltage waveform data at the one or more sensor units on the second side of the event location, when the one or more sensor units on the second side are in the sending mode. The control unit is configured to determine the event location where the broken rail event has occurred, in use, between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, by detecting elevated voltage waveform data of the plurality of sensor units, with a relative elevation of the voltage waveform data decreasing, in send mode and receive mode, with increasing distance from the one of the opposed ends. The control unit is configured to determine that the broken rail event corresponds to a loose joint event by detecting, after initial detection of the broken rail event, relatively less elevation and, if previously detected, attenuation, of voltage waveform data. The control unit is configured to determine that the broken rail event corresponds to the loose joint event by detecting, after initial detection of the broken rail event, relatively less elevation and, if previously detected, attenuation, of voltage waveform data during at least neutral zone temperature transitions. The control unit is configured to determine an event location where a train present event has occurred, in use, at a sensor unit, of the plurality of sensor units, or between adjacent sensor units, of the plurality of sensor units, by detecting a relatively largest attenuation in voltage waveform data of the sensor unit or adjacent sensor units, when in send mode and receive mode. The control unit is configured to determine an event location where a train present event has occurred, in use, at a sensor unit, of the plurality of sensor units, or between adjacent sensor units, of the plurality of sensor units, by detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from the sensor unit or adjacent sensor units in send mode and receive mode. The control unit is configured to determine the existence of an event, comprising one or more of a broken rail event, a train present event, or a loose joint event, that has occurred, by detecting one or more of an attenuation or elevation of voltage waveform data at one or more sensor units, of the plurality of sensor units, beyond a predetermined threshold change. The predetermined threshold change corresponds to a 5% elevation or greater for determining the existence of a broken rail event or a loose joint event detected by a sensor unit, of the plurality of sensor units, in send mode. The predetermined threshold change corresponds to a 10% elevation or greater, or a 20% attenuation or greater, for determining the existence of the event by a sensor unit, of theplurality of sensor units, in receive mode. The control unit is configured to determine calibrated voltage waveform data for one or more of the sensor units, of the plurality of sensor units, based on rolling averages when in the normal state. The section of railroad track comprises a switch, and one or more of: the switch has shunts on either side of the switch and forms a blind zone in the section; all switch joints intended to pass current through the switch are bonded; or insulated joints are installed or present on the switch to prevent the switch from shunting the rails in operation. The rail-to-rail shunts comprise one or more of un-isolated switches, or specifically placed boundary shunts. The control unit comprises a plurality of control units. The control unit is connected to send instructions and receive data wirelessly from the plurality of sensor units. The control unit comprises a remote server that is accessible by the plurality of sensor units via the internet. The plurality of sensor units are distributed along the railroad at even intervals. The plurality of sensor units comprises three or more sensor units. The plurality of sensor units are configured to produce the stimulus electrical voltage by direct current (DC). The control unit is configured to reduce alternating current (AC) interference from voltage waveform data measured by the plurality of sensor units. The control unit is configured to reduce AC interference from the voltage waveform data by shifting, and overlaying the voltage waveform data on itself, by an interval of (n + ‘A)* wavelength of the AC-induced interference, where n is an integer, and the wavelength is of the AC interference. The AC interference is present at two or more frequencies. The control unit is configured to reduce the AC interference from the voltage waveform data by shifting, and overlaying the voltage waveform data on itself, by an interval that is selected to be equal to (ni + A )*periodi = (nm+ A)* periodm, where n is an integer, 1 to m denote respective AC interferences of different frequencies, and the period is the time value of a single period or wavelength of the respective AC interference. The plurality of sensor units are configured to, in the sending mode, produce a stimulus as a plurality of pulses spaced by the interval. Determining further comprises determining an event location, being a length of the section of railroad track between adjacent sensor units or between one of the opposed ends of the section of railroad track and an adjacent sensor unit, where a broken rail event or a train present event has occurred. Determining further comprises determining the event location where a broken rail event has occurred, in use, between adjacent sensor units by detecting: elevated voltage waveform data at the sensor units on a first side of the event location, and attenuated voltage waveform data at the sensor units on a second side of the event location, when the sensor units on the first side are in the sending mode; and attenuated voltage waveform data at the sensor units on the first side of the event location, and elevated voltage waveform data at the sensor units on the second side of the event location, when the sensor units on the second side are in the sending mode. Determining comprises determining the event location where a broken rail event has occurred, in use, between one of the opposed ends of the section of railroad track and an adjacent sensor unit, by detecting elevated voltage waveform data of the plurality of sensor units, with a relative elevation of the voltage waveform data decreasing, in send mode and receive mode, with increasing distance from the adjacent sensor unit. Determining comprises determining the event location where a train present event has occurred, in use, at an adjacent sensor unit or between adjacent sensor units by detecting attenuated voltage waveform data of the plurality of sensor units in send mode and receive mode. Determining comprises determining the event location where a train present event has occurred, in use, between adjacent sensor units by detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from the adjacent sensor units in send mode and receive mode. Determining comprises determining the event location where a train present event hasoccurred, in use, at an adjacent sensor unit by: detecting a relative local maximum attenuation of voltage waveform data at the adjacent sensor unit in send mode; and detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from sensor units adjacent to the adjacent sensor unit. Determining further comprises determining an event location, being a length of the section of railroad track between two or more sensor units, of the plurality of sensor units, or between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, where a broken rail event, a train present event, or a loose joint event has occurred. Determining further comprises determining the event location where a broken rail event has occurred, in use, between sensor units, of the plurality of sensor units, by detecting: elevated voltage waveform data at one or more of the sensor units, of the plurality of sensor units, on a first side of the event location, and attenuated voltage waveform data at one or more of the sensor units, of the plurality of sensor units, on a second side of the event location, when the one or more sensor units on the first side are in the sending mode; and attenuated voltage waveform data at the one or more sensor units on the first side of the event location, and elevated voltage waveform data at the one or more sensor units on the second side of the event location, when the one or more sensor units on the second side are in the sending mode. Determining comprises determining the event location where the broken rail event has occurred, in use, between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, by detecting elevated voltage waveform data of the plurality of sensor units, with a relative elevation of the voltage waveform data decreasing, in send mode and receive mode, with increasing distance from the one of the opposed ends. Determining comprises determining that the broken rail event corresponds to a loose joint event by detecting, after initial detection of the broken rail event, relatively less elevation, and, if previously detected, attenuation, of voltage waveform data. Determining comprises determining the event location where a train present event has occurred, in use, at a sensor unit, of the plurality of sensor units, or between adjacent sensor units, of the plurality of sensor units, by detecting a relatively largest attenuation in voltage waveform data of the sensor unit or adjacent sensor units, when in send mode and receive mode. Determining comprises determining the event location where a train present event has occurred, in use, between adjacent sensor units, of the plurality of sensor units, by detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from the adjacent sensor units in send mode and receive mode. Determining comprises processing voltage waveform data to reduce alternating current (AC) interference. The plurality of nodes comprise one or more gateway routers. The one or more gateway routers are characterized by superiority relative to the other of the plurality of nodes, of one or more of battery performance, long-range link quality, power source capacity, power source permanence, and a wired connection to the long-range network. In a first mode, each of the plurality of nodes assign and use one or more gateway routers as the border router in preference to the other of the plurality of nodes. In a second mode, each of the plurality of nodes assign and use another of the plurality of nodes as a border router when a ranking of the gateway router drops below a predetermined threshold or a ranking of the another of the plurality of nodes. Each of the plurality of nodes are configured to dynamically assign a border router for the node based on rankings dynamically assigned to each of the plurality of nodes. A ranking of each node is based on one or more of the following performance characteristics: a battery capacity, battery performance, available charging capacity, long-range link quality, microcontroller unit (MCU) utilization, number of other nodes in communication range with the node, number of other nearby candidateborder routers, distance to the nearest suitable candidate border router, long-range network load, and short-range bandwidth. The ranking of each node is based on at least battery capacity, available charging capacity, and long-range link quality. Each of the plurality of nodes are configured to broadcast, to other of the plurality of nodes, performance characteristics of the node at periodic intervals or because of a change in performance characteristics. Each of the plurality of nodes are configured to self-rank and broadcast a self-ranking for the node to other of the plurality of nodes. Each of the plurality of nodes are configured to dynamically assign and use one or more of the plurality of nodes as a border router based on the rankings provided by each of the plurality of nodes and a short-range link quality to each of the plurality of nodes. There are a plurality of border routers available for a node; and the plurality of nodes are configured to failover to other of the plurality of border routers when a first of the plurality of border routers is instructed, and fails, to send a long-range transmission through the long-range network. In the event of a border router being instructed, and failing, to send data through the long-range network, one or more of the plurality of nodes storing and subsequently re-attempting to send or instruct the sending of the data. The power source of one or more of the plurality of nodes comprises a low-watt power source. The power source has an average energy production capacity of 15 W / hr or less daily. The power source comprises a solar panel. Each of the plurality of nodes comprises a housing and is mounted above-ground on a pole. The long-range wireless network comprises one or more of the internet, a cellular network, and a satellite network. The geographical area comprises a transport corridor. The transport corridor comprises a railway track for a train. The plurality of nodes are configured to maintain a train on the railway track in continuous communication with the long-range wireless network. The geographical area comprises a remote geographical area with portions or the entirety of which being out of contact with a cellular network other than via one or more border routers. The long-range transceiver comprises one or more of: a cellular transceiver, a satellite transceiver, a broadcast radio transceiver, and a microwave transceiver. The short-range transceiver comprises one or more of a low frequency radio transceiver, an infrared transceiver, a Bluetooth transceiver, a Wi-Fi transceiver, and a mesh network transceiver. The short-range transceiver comprises a low frequency radio transceiver, such as a 2.4 GHz or 915 MHz wireless transceiver. A back-end system comprising a server processor connected to receive and transmit communications from and to, respectively, the plurality of nodes via the long-range network. The server processor is connected to serve and relay communications information from the plurality of nodes to third parties via the long- range network. A plurality of sensors connected to provide information to the plurality of nodes. The plurality of sensors are configured to sense and provide information on one or more of motion, temperature, vibration, tilt, rail or corridor integrity, seismic activity, humidity, water levels, weather, flooding, proximity, obstacle, air quality factors, wildlife, sound, and visual elements. The plurality of nodes comprise one or more gateway routers; in a first mode, each of the plurality of nodes assign and use one or more gateway routers as the border router in preference to the other of the plurality of nodes; and in a second mode, each of the plurality of nodes assign and use another of the plurality of nodes as a border router when a ranking of the gateway router drops below a predetermined threshold or a ranking of the another of the plurality of nodes. Each of the plurality of nodes are configured to dynamically assign a border router for the node based on rankings dynamically assigned to each of the plurality of nodes. Each of the plurality of nodes broadcasting, to other of the plurality of nodes, performance characteristics of the node at periodic intervals or because of a change in performance characteristics. Each of the plurality of nodes self-ranking itself and broadcasting a self-ranking for the node to other of the plurality of nodes. At one of the plurality of nodes: assigningone of the other of the plurality of nodes as a border router based on the rankings provided by each of the plurality of nodes and a short-range link quality to each of the plurality of nodes; and transmitting a message to the border router. There are a plurality of border routers available for a node, and further comprising: attempting to transmit data by one of the plurality of border routers; and in the event of failure, attempting to transmit data by subsequent of the plurality of border routers. In the event of no other suitable border router being located or all possible border routers failing to transmit the data, storing and subsequently re-attempting to send or instruct the sending of the data. The plurality of nodes cooperate to collectively assign rankings to each node. The plurality of nodes collectively form a self-healing wireless mesh network. The low-watt power source has a wattage of 50 watts or less. Transmitting a message off- network through the long-range network via a border router.

[0015] The foregoing summary is not intended to summarize each potential embodiment or every aspect of the subject matter of the present disclosure. These and other aspects of the device and method are set out in the claims. BRIEF DESCRIPTION OF THE FIGURES

[0016] Embodiments will now be described with reference to the figures, in which like reference characters denote like elements, by way of example, and in which:

[0017] Fig. 1 is a schematic view of a railroad track sensor system with a plurality of sensor units, the plurality of sensor units configured to cycle through a rotation where each sensor unit is set to operate, one at a time, in a sending mode, to determine whether the section of the railroad is in a normal state, a broken rail state, or a train present state. Fig. 2 is a graphical representation of profiles of voltage waveform data obtained from the plurality of sensor units of Fig. 1, depicting the differences in waveform data based on which sensor unit is acting in sending mode. Fig. 3 is a schematic view of the railroad track sensor system of Fig. 1, illustrating operation during a rail break between two sensor units. Fig. 4 is a graphical representation of profiles of voltage waveform data obtained from the plurality of sensor units of Fig. 3, depicting the differences in waveform data based on which sensor unit is acting in sending mode. Fig. 5 is a schematic view of the railroad track sensor system of Fig.1, illustrating operation during a rail break between a sensor unit and a shunt. Fig. 6 is a graphical representation of profiles of voltage waveform data obtained from the plurality of sensor units of Fig. 5, depicting the differences in waveform data based on which sensor unit is acting in sending mode. Fig. 7 is a schematic view of the railroad track sensor system of Fig. 1, illustrating operation under normal rail conditions with a train present. Fig. 8 is a graphical representation of profiles of voltage waveform data obtained from the plurality of sensor units of Fig. 7, depicting the differences in waveform data based on which sensor unit is acting in sending mode. Fig. 9 is a schematic view of a communications system, illustrating a track-side wireless mesh network communicating with a locomotive in a remote geographical area with gaps in cellular coverage. Fig. 10 is a plot of relative numerical representations of voltage (a voltage waveform / waveform data) detected by a sensor unit over time while the sensor unit carries out a stimulus electrical voltage, under a normal state, a broken rail state, and a train present state. Figs. 11-13 are plots of the relative numerical representations of voltage detected by a plurality (three in this case) sensor units, over time, with each data point in the plot having been calculated, for each sensor unit, by taking an average of the relative numerical representations of a voltage waveform detected by the sensor unit during a stimulus electrical voltage event, with the chart depicted around 9:00 a broken rail event, and train present events at around 4:00, 14:00, 15:30, 20:00, and 22:00. Fig. 14 is a plot of relative numericalrepresentations of voltage (a voltage waveform / waveform data) detected by a sensor unit over time, illustrating a baseline AC -induced interference signal that might be present on a railroad track as a result of an AC current being carried in power lines adjacent the railroad track. Fig. 15 is a plot of relative numerical representations of voltage (a voltage waveform / waveform data) detected by a sensor unit over time, while the sensor unit carries out a stimulus electrical voltage in the form of a pair of pulses, in an idealized scenario without AC -induced interference. Fig. 16 is a plot that includes both the voltage waveform data of Figs. 14 and 15, in a non-additive fashion. Fig. 17 is a plot that overlays the voltage waveform data of Figs. 14 and 15, to illustrate the impact of AC-induced interference on measured electrical signal (resultant voltage) at a sensor unit. In addition, the voltage waveform data shown is taken from a sensor unit that uses reverse voltage circuitry protection. Fig. 18 is a plot that includes both the plot of Fig. 17 and a duplicate plot that is shifted by 4 and a half wavelengths of the AC interference signal. Fig. 19 is a plot that overlays the two plots of Fig. 18, to illustrate processed voltage waveform data with reduced effect of AC-induced interference on the railroad track. Fig. 20 is a plot of relative numerical representations of voltage (a voltage waveform / waveform data) detected by a sensor unit over time, illustrating a pair of baseline AC -induced interference signals that might be present on a railroad track, at two different respective frequencies, namely at fi = 60 Hz and f2 = 86.7 Hz in the example. Fig. 21 is a plot that includes both the voltage waveform data of Fig. 20, and that of a pair of pulses, in a non-additive fashion. Fig. 22 is a plot that overlays the voltage waveform data of Fig. 21, to illustrate the impact of AC -induced interference on measured electrical signal (resultant voltage) at a sensor unit by a pair of AC interference types. In addition, the voltage waveform data shown is taken from a sensor unit that uses reverse voltage circuitry protection. Fig. 23 is a plot that includes both the plot of Fig. 22 and a duplicate plot that is shifted. Fig. 24 is a plot that overlays the voltage waveforms of Fig. 23, to at least partially cancel out the AC interference of the two frequencies.DETAILED DESCRIPTION

[0018] Immaterial modifications may be made to the embodiments described here without departing from what is covered by the claims.

[0019] A railway is a form of transportation that utilizes a network of tracks to transport passengers and goods over long distances. Railways have been an important mode of transportation since the 19th century, revolutionizing the way people and goods move across countries and continents. Rail transport offers several advantages over other forms of transport, such as being relatively safe, efficient, and cost-effective. Railways can carry large volumes of goods and people and can also be used to transport bulky or heavy items that may be difficult to move by road or air. Railways may also have a lower environmental impact compared to other modes of transportation, such as cars and airplanes, as rail transportation consumes about four times less energy per ton-km or per passenger-km than road transportation.

[0020] Maintaining and operating a functional rail network is a challenging process. A railway may have lines that extend 100s of kilometers into remote areas, in dynamic environments. A railway, like any form of infrastructure, is subject to deterioration over time, whether from wear and tear, weather, or other environmental factors such as rockslides, foundation shift, and in some cases, wildlife. The occurrence of rail breaks is a natural result of these and other factors. A rail break, if left undetected, can lead to everything from minor delay tocatastrophic derailment, which can be devastating and result in damage and loss of life. A rail break may occur in subtle fashion, such as when a train passes over a section of track, safely navigating the section but leaving a broken rail behind, which may cause a derailment of a subsequent train. Thus, in the proper maintenance and operation of a railroad, it is crucial to be able to immediately detect and respond to (i.e., fix) a rail break, whether large or small.

[0021] Monitoring remote sections of railways can be a challenging task for various reasons. The main challenge is usually the sheer size of most railway networks, which may stretch across vast distances and difficult terrain. The size of remote railway networks can make it difficult for personnel to access these remote areas regularly, making it harder to monitor the tracks and surrounding areas for any signs of damage, such as breaks, theft, or other incidents. Remote sections of railways may lack adequate communication infrastructure, making it difficult to relay real-time information to a central command center. Deploying railway sensors, drones, or other monitoring devices can offer real-time information about the condition of the tracks and surrounding areas, thereby enhancing the monitoring and maintenance of remote railway networks. Remote monitoring and maintenance can also be improved through the use of satellite communication and other advanced technologies.

[0022] A conventional railway system employs a rail track as a part of a signal transmission path to detect the existence of either a train or a rail break in a block section. In such a method, the track is electrically divided into a plurality of sections, each having a predetermined length. Each section forms a part of an electric circuit and may be referred to as a track circuit. A basic track circuit used on railroads for years to indicate the condition or occupancy of a track section or block utilizes a voltage source, such as a transmitter device, at one end of a track section and a voltage detection device, such as a relay or a receiver device, at the other end of the section. A transmitter device and a receiver device are arranged respectively at either ends of the track circuit. The transmitter device transmits a signal for detecting a train or rail break continuously or at variable intervals and the receiver device receives the transmitted signal. If a train or rail break is not present in the section formed by the track circuit, the receiver receives the signal transmitted by the transmitter. If a train or rail break is present, the receiver receives a modified signal transmitted by the transmitter, because of the change in the electrical circuit formed by the track and break, or track and train. In general, train presence modifies the track circuit through the addition of a shunt resistance from rail to rail. Break presence modifies the circuit through the addition of an increased resistance in the rail. Break or train detection is generally accomplished through a comparison of the signal received with a threshold value. Current technology now additionally provides means for bidirectionally coding this energy to transmit and receive information through the rails, as well as track occupancy detection by the shunting action of the wheels of a train. These systems provide block occupancy information at both ends of the track circuit, as well as communicating occupancy in general through several track sections to a control point where the information may be transmitted to a central office for display.

[0023] Indication of an occupied block is provided by the rail-to-rail shunt between the wheels of an entering or already present train, which establishes a low resistance path and thus a loss of signal strength at the receiver at each end of the block. Indication of the track circuit as occupied is the safe state for the track circuit under failure, i.e., if the track circuit experiences some form of failure, the same indication is given as an occupied track to prevent rail vehicles from operating at high speed in that track section. A break in the rails also provides the same indication as an occupied track or a failure of one or more of the track circuit components. A practical problem,however, arises from this lack of distinction between types of failures. The railroad personnel responsible for track circuit maintenance and rail repair are typically different persons belonging to different groups. Current practice is for track circuit maintenance personnel to isolate a track circuit failure, then contact the personnel responsible for rail repair if the rail is broken. A broken rail, therefore, can result in a long delay in rail traffic while one and then another maintenance crew is dispatched to effect repairs.

[0024] Conventional track circuits are generally applied to blocks of about 2 to 2.5 miles in length for detecting a train. In such a block, a train should exhibit a train shunt resistance of about 0.06 ohms or less, and the ballast resistance or the resistance between the independent rails will generally be greater than about 3 ohms / 1000 feet. As the block length becomes longer, the overall resistance of a track circuit decreases due to the parallel addition of ballast resistance between the rails. Through this addition of parallel current paths, additional current flows through the ballast and ties and proportionally less through the receiver. Thus, the signal to noise ratio of the track circuits degrades with longer block lengths. To properly maintain a track circuit, periodic tests are required to assure the equipment's capability to detect a shunted track, as well as, in many cases today, testing the track circuit relay or receiver for proper level of operation. These tests are currently performed manually at six month and two-year intervals, respectively, by having track circuit maintenance personnel visit each site with appropriate test equipment.

[0025] Referring to Figs. 1, 3, 5, 7 and 9, a railroad track sensor system 10 is disclosed. The track sensor system 10 comprises a plurality of sensor units 28. The plurality of sensor units 28 may be distributed at various respective locations along a section 12 of railroad track 14. The plurality of sensor units 28 may be distributed along the railroad track 14 at even intervals, although the selection of interval size may be different for each interval and may be based on geography, convenience, resources, or other factors. The status of the section 12 of the railroad track 14 may be assessed with the plurality of sensor units 28. The section 12 of railroad track 14, may comprise a pair of opposed parallel rails 15 bounded on opposed ends 11 by rail-to-rail shunts 16, a component, such as a bonded part, of a switch acting as a shunt 18 and ballast 50. The track 14 may include railroad ties or sleepers 48. The plurality of sensor units 28 may be configured to cycle through a rotation where each sensor unit 28 is set to operate, one at a time, in a sending mode. When a sensor unit 28 is in the sending mode, the sensor unit 28 may produce a stimulus electrical voltage across the parallel rails 15. The plurality of sensor units 28 may be set to simultaneously operate in a receiving mode, wherein the plurality of sensor units 28 may each measure a resultant voltage across the parallel rails 15 of the railroad track 14. The track sensor system 10 may comprise a control unit 78. The control unit 78 may be configured to compare voltage waveform data, such as is visualized in the figure as a plurality of profiles 40 of the voltage waveform data shown, with the resultant voltages being derived from waveform data, measured by the plurality of sensor units 28 during the rotation, with calibrated voltage waveform data, such as visualized in the figure as a plurality of profiles 38 of calibrated voltage waveform data. Comparison of the voltage waveform data to the calibrated voltage waveform data is used to determine whether the section 12 of the railroad track 14 is in a normal state, a broken rail state, or a train present state.

[0026] Referring to Figs. 1-9, the track sensor system 10 may be used to detect the presence of a break 20 in one or both of the opposed rails 15 of the railroad track 15, which may be indicated by a loss of continuity in a closed loop of the railroad track 14. The system 10 may comprise either a single sensor, for short blocks of railroad track 14,or a plurality of sensor units 28. A plurality of sensor units 28 may be located at suitable intervals, such as ranging from a half a mile to 2 miles, although intervals greater or smaller than this range may be used. The boundaries of the section 12 of railroad track 14 may be defined by shunts 16 across the opposed rails 15. The shunts 16 may be normally occurring shunts 16 such as un-isolated switches, or specifically placed boundary shunts 16. The system 10 may provide a stimulus voltage across the adjacent opposed rails 15 and measure the resultant voltage waveform data created by the impedance of the railroad track 14 at the sensor unit 28 which is functioning in a sending mode. If more than one sensor unit 28 is located on the section 12 of track 14, the plurality of sensor units 28 will each take a turn functioning in the sending mode while the other units measure the rail-to-rail waveform data at each of their individual locations. The units may rotate through the sending mode role once per update interval (Figs. 2, 4, 6, 8). The waveform data from the plurality of sensor units 28 may be sent to a server, such as control unit 78 (for example a back-end system). One or more control units may be used, for example each sensor unit 28 may have a control unit that communicates with the control units of other sensor units 28. The control unit may monitor for changes in the each of the related waveform data measured at the fixed intervals along the track 14. When a plurality of sensor units 28 are present in a section 12 of track 14, the waveform data from the full sending mode role rotation may be aggregated to provide redundancy and improved detection results. The primary conclusions that may be inferred from the waveform data include a normal state (Figs. 1 and 2 - no break detected), a broken rail state (Figs. 3-6) and a train present state (Fig. 7-8). The location of the rail break 20 may be detected as being, between any two sensor units 28 in the section 12 of track 14 (Figs. 3-4), outside of a range defined by any one or two sensor units 28 in the section 12 of track 14, between a sensor unit 28 and an end of the shunted section of track (Figs. 5-6) or, near a sensor unit 28 in the section 12 of track 14. Other diagnostic data may be collected as well, for example, whether the leads are connected to the rails, shorted leads, interfering electrical noise and other date may be collected.

[0027] Referring to Figs. 1, 3, 5, 7, and 9, the section 12 of railroad track 14 may be bounded on opposed ends 11 by rail-to-rail shunts 16. A shunt 16 may be an electrical connection that allows electrical current to flow around the section 12 of the railroad track 14 with relatively minimal resistance. For example, a conductive wire may be electrically connected between the two rails. Shunts 16 may be used in the railroad tracks 14 to detect the presence of a train or a break in a particular section of track 14. The rail-to-rail shunts 16 that may be used may comprise one or more of un-isolated switches, or specifically placed boundary shunts. Pre-existing or naturally occurring shunts may be leveraged to define the ends of the section of track, such as provided by other infrastructure of or associated with the railway.

[0028] Referring to Figs. 2, 4, 6, 8, and 9 system 10 comprises the control unit 78. The control unit 78 may comprise a plurality of control units 78. The control unit 78 may be connected to send instructions and receive data wirelessly from the plurality of sensor units 28. In other cases, wired connections may be used. The control unit 78 may comprise a remote processor that is accessible by the plurality of sensor units 28 via the internet or other remote communications protocol methodology. The control unit 78 may receive and analyze waveform data from the plurality of sensor units 28. The control unit 78 may analyze the waveform data to determine whether the section of railroad track 14 is in in a normal state, a broken rail state, or a train present state. In some cases, plural control units 78 may each perform respective analyses on the data, such as the data from individual or grouped sensor units 28, and the analyzed data may be shared with other nodes (units 78 or sensor units 28), bused and sent to a further control unit78, or provided directly or indirectly to one or more command control units 78. The control units 78 may be adjacent or remote relative to the system 10.

[0029] Referring to Figs. 1-9, the plurality of sensor units 28 may be configured to measure voltage that can be used to produce waveform data. As described below with respect to Figs. 10-13, waveform data may be processed using the waveform itself, or by processing waveform data measured or calculated from the waveform or other voltage data itself. A waveform generally refers to a mathematical representation of a wave, especially a graph obtained by plotting a characteristic of the wave against time. During the send, the plurality of sensor units 28 all simultaneously operate in a receiving mode, and each measure the resultant voltage generated by the sensor unit 28 functioning in the sending mode. The waveform data is received and analyzed by the control unit 78 or units 78 if there are plural units 78. Referring to Fig. 1, example waveform data is illustrated, with each number in the time slot y-axis referring to the number of the sensor unit 28 in Fig. 1 being in the sending mode for the respective data. The x- axis on the other hand, illustrates a waveform that is created by the discrete resultant voltages measured by each sensor unit 28 in the receiving mode, with dashed lines used to identify the discrete points of data (resultant voltages) measured by respective sensor units 28. The waveforms are shown with each sensor unit 28 in order according to the order of units 28 along the railroad track. The plurality of sensor units 28 may rotate through the role of functioning in the sending mode, and the waveform data from the full sending mode role rotation may be aggregated to provide redundancy and improved detection results. Each sensor unit 28 may also operate in the receiving mode during sending, as is shown in the data of Fig. 2.

[0030] The data in Fig. 2 represents example normal or calibrated voltage waveform data, that is, resultant voltages measured through a full cycle (with a full cycle achieved by having each sensor unit 28 individually cycle through the sending mode). A calibrated waveform may represent a normal state for a particular section of railroad track. In a normal state, the track is neither broken, nor unintentionally shunted between ends, nor interrupted by the presence of a train. The calibrated waveform data may need to be periodically obtained and updated as sensor data is expected to drift over time and in different weather conditions. In some cases, the system may not use a fixed calibrated waveform. The devices or units may simply report the waveform data back to a server. The server itself may not apply a fixed preset calibrated waveform. The server may look for deviations from a baseline that adjusts to the particular electrical properties of the track section that each unit is located in. The baseline data may be given an initial calibration and then dynamically adjusts as the track section’s electrical properties change do to environmental factors, such as, rain, snow, freezing ground, presence of ion-producing chemical compounds, and other factors. The calibrated voltage may be a rolling average, for example obtained via a rolling average algorithm. If there is a significant change from the rolling averages of one or more sensor units, then an alert may be published. This cycle may be repeated and replicated throughout the system. Referring to Figs. 1-8, profdes 40 waveform data may be compared with profiles 38 of normal or calibrated voltage waveform data to determine whether the section 12 of the railroad track 14 is in a normal state (Figs. 1-2), a broken rail state (Figs. 3-6), or a train present state (Figs. 7-8). In the calibrated waveform data, the resultant voltages would be expected to follow a predictable pattern where the highest resultant voltages are measured at the sending unit 28, with a gradual decline in non-zero resultant voltages at a predetermined rate over sensor units 28 with increasing distance from the sending unit 28 in both directions. The rate of decline may be measured in analyzing the waveform data, in order to determine a natural loss of current perinterval of length between units 28 or per unit length of track 14. The plurality of sensor units 28 may be configured to produce a stimulus electrical voltage by a suitable method, such as using a direct current (DC). In other cases, alternating current (AC) or other suitable configurations may be used. The plurality of sensor units 28 may comprise three or more sensor units 28, such as five or more, which may assist in compartmentalizing the monitoring of the track 14 into smaller lengths for more accurate event locating (discussed below). Referring to Figs. 1-9, the track sensor system 10 may detect an event location 21, namely the location in the section 12 of the railroad track 14 where an event, such as a broken track or the presence of a train, is occurring. The event location 21 may correspond to the location of one or more of a break 20 in one or both of the opposed rails 15 of the railroad track 14 or a train 22 present in the section 12 of the railroad track 14. The control unit 78 may be configured to determine the event location 21, being a length of the section 12 of railroad track 14 between adjacent sensor units 28 (sensor units that are adjacent one another and that are closest to the event location) or at one or more of the sensor units 28, or otherwise between one of the opposed ends 11 of the section 12 of railroad track 14 and an adjacent sensor unit 28 (a sensor unit adjacent the one of the opposed ends), where a broken rail event or a train present event has occurred.

[0031] Referring to Figs. 3-4, the system may be configured to determine an event location 21 where a broken rail event has occurred, in use, on the section 12 of track 14. When a rail break 20 is present on the section 12 of railroad track 14, a broken rail event may be detected by analysis of the waveform data. The control units or unit 78 may compare the voltage waveform data such as is visualized in the figure as a plurality of profiles 40 of the voltage waveform data shown, with the resultant voltages being derived from waveform data measured by the plurality of sensor units 28 during the rotation (Fig. 4), with calibrated voltage waveform data (Fig. 2), such as visualized in the figure as a plurality of profdes 38 of calibrated voltage waveform data, to determine where a broken rail event has occurred. The control unit 78 may determine that a broken rail 21 has occurred between two or more sensor units 28, such as adjacent sensor units 28, by a) detecting elevated voltage waveform data (illustrated by the elevated parts 45 of profiles 40) at the sensor units 28 on a first side 24 of the event location 21, and attenuated voltage waveform data (illustrated by the attenuated parts 42 of profiles 40) at the sensor units 28 on a second side 26 of the event location 21, when one of the sensor units 28 on the first side 24 is in the sending mode (the top two waveforms in Fig. 4), and b) detecting attenuated voltage waveform data at the sensor units 28 on the first side 24 of the event location 21, and elevated voltage waveform data at the sensor units 28 on the second side 26 of the event location 21, when one of the sensor units 28 on the second side 26 is in the sending mode (the bottom three waveforms in Fig. 4). In the Figures shown, a change 41, in voltage waveform from calibrated data, at a sending sensor unit is identified as reference numeral 41, while a change 43, in voltage waveform from calibrated data, at a receiving sensor unit is identified as reference numeral 43. The control unit 78 may recognize such a pattern by comparing the measured waveform data, for example the changes 41 and 43, with the calibrated waveform data. The relative elevation of signal on the sending side of the break may be attributed to the additional current that would normally travel to the other side of the break but is forced to travel to one side by the break itself. The relative attenuation of signal on the non-sending side of the break may be attributed to the fact that little to no signal is passed to the second side of the break due to the natural resistance to current of the ballast or other environment between the separated ends of the broken rail. The elevated voltage waveform data and the attenuated voltage waveform data may be elevated and attenuated, respectively, compared to the calibrated waveform data. In the event that a break somehow spans a sending sensor unit 28, suchthat the break or plural breaks interrupt current flow in the rails up and downstream of the sensor unit 28, the above pattern may still be recognized by sending mode results from sensor units 28 on either side of the break or breaks. When the sensor unit 28 that is isolated on both sides by the breaks is in sending mode, that sensor unit 28 may show elevated voltage at that sensor unit 28 and attenuated data at all other sensor units 28, whether on the first or second side of the break. The remaining sensor units 28 would generate voltage waveform data similar to that shown in Fig. 4 in all other sending modes of the cycle.

[0032] Referring to Figs. 5 and 6, the control unit 78 of the railroad track sensor system 10 may be configured to determine an event location 21 where a broken rail event has occurred, in use, between one of the opposed ends 11 of the section 12 of railroad track 14 and another sensor unit, such as an adjacent sensor unit 28 (adjacent to the end 11). When a rail break 20 is present on the section 12 of railroad track 14, a broken rail event may be detected by the control unit or units 78. The control unit 78 may compare the voltage waveform data, such as is visualized in the figure as a plurality of profiles 40 of the voltage waveform data shown, with the resultant voltages being derived from waveform data, measured by the plurality of sensor units 28 during the rotation, with calibrated voltage waveform data, such as visualized in the figure as a plurality of profiles 38 of calibrated voltage waveform data, to determine where a broken rail event has occurred. The control unit 78 may be configured to determine an event location 21 where a broken rail event has occurred between one of the opposed ends 11 of the section 12 of railroad track 14 and an adjacent sensor unit 28, by detecting elevated voltage waveform data (elevated parts 45) of the plurality of sensor units 28, with a relative elevation of the voltage waveform data decreasing, in send mode and receive mode, with increasing distance from the one of the opposed ends, and in some cases from the adjacent sensor unit 28. The relative elevation of the waveform data can be attributed to the fact that current near the end 11 of the track 14 at which the break 20 has occurred can only effectively travel down the track 14 toward the sensor units 28, as the break 20 limits the amount of current that can be shunted across the rail by the shunt 16.

[0033] Referring to Figs. 7-8, the control unit 78 of the railroad track sensor system 10 may be configured to determine an event location 21 corresponding to the location of a train (a train present event). When a train 22 is present on the section 12 of railroad track 14, a train present event may be detected by the control units or unit 78. The control unit 78 may compare the voltage waveform data, such as is visualized in the figure as a plurality of profiles 40 of the voltage waveform data shown, with the resultant voltages being derived from waveform data, measured by the plurality of sensor units 28 during the rotation, with calibrated voltage waveform data, such as visualized in the figure as a plurality of profiles 38 of calibrated voltage waveform data, to determine where a train present event has occurred. The control unit 78 may be configured to determine an event location 21 where a train present event has occurred at an adjacent sensor unit 28 (a sensor unit adjacent to the event location) or between adjacent sensor units 28 by detecting attenuated voltage waveform data (attenuated parts 42 of profiles 40) of the plurality of sensor units 28 in send mode and receive mode. The relative attenuation may be attributed to the fact that the train 22 itself acts as a dynamic shunt that reroutes current across the opposed rails of the track 14. A relative maximum attenuation may indicate the location of the train. The relative attenuation may diminish with increasing distance from the train 22, and thus in the example shown, sensor unit 5 shows the least amount of relative attenuation. By contrast, the relative attenuation is the highest at the sensor unit 28 (unit 2 in Fig. 7) at which the train 22 is nearest or is directly overhead, at least when that sensor unit 28 is in the sending mode. More specifically, the control unit 78 may be configured todetermine an event location 21 where a train present event has occurred at an adjacent sensor unit 28 by detecting a relative local maximum attenuation of voltage waveform data at the adjacent sensor unit 28 in send mode and detecting attenuated voltage waveform data of the plurality of sensor units 28, with a relative attenuation of the voltage waveform data decreasing with increasing distance from sensor units 28 adjacent to the adjacent sensor unit 28 in which the train present event has occurred. In some cases, the train 22 may be present between adjacent sensor units 28 - in such a case, the sensor unit 28 that is closest to the train 22 ought to show the most relative attenuation of signal when in the sending mode, however, relative attenuation ought to be smaller at such a unit 12 than in the example shown where the train 22 is directly overhead that sensor unit 28. The control unit 78 may be configured to determine an event location 21 where a train present event has occurred between one of the opposed ends 11 of the section 12 of railroad track 14 and the adjacent sensor unit 28 by detecting a relative local maximum attenuation of voltage waveform data at the end of section sensor unit 28 that is adjacent to the opposed end 11 where the train is present, at least when that sensor unit 28 is in the send mode. A relative attenuation of the voltage waveform data may decrease with increasing distance from train present event.

[0034] Referring to Figs. 4, 6, and 8, the control unit may be configured to determine the existence of an event that has occurred, by detecting one or more of an attenuation or elevation of voltage waveform data (changes 41 and 43) at one or more sensor units, of the plurality of sensor units, beyond a predetermined threshold change. Referring to Figs. 4 and 6, the predetermined threshold change may correspond to a 5%, for example 7%, elevation or greater for determining the existence of a broken rail event or a loose joint event detected by a sensor unit, of the plurality of sensor units, in send mode. As shown in Figs. 4 and 6, the changes 41 in sending mode sensor units closest to the break meet the threshold, while some of the more distant sensor units will not. The ability for the system to respond to a change in data allows the system to operate even in the event one or more sensor units fails to properly function. Thus, in Fig. 4, if sensor units 2 and 3 fail, sensor units 1 and 4 may still detect the break if the threshold change is set high enough. Similarly, the predetermined threshold change may correspond to a 10% elevation or greater, or a 20% attenuation or greater, (changes 43) for determining the existence of the event by a sensor unit, of the plurality of sensor units, in receive mode. The ability to use receive mode sensor units in such a fashion also improves the reliability of the system in the event that one or more sensor units fails to properly function. As above, the control unit is configured to determine calibrated voltage waveform data for one or more of the sensor units, of the plurality of sensor units, based on rolling averages when in the normal state.

[0035] Referring to Fig. 9, the sensor system 10 may operate in a mesh or other suitable network. Remote geographical areas exist where gaps in cellular or satellite networks (or other long-range networks) make long-range communications difficult or impossible. Many industrial operations, such as rail, mining, pipeline, harvesting, and others, are carried out in remote geographical areas, such as remote forests, mountain ranges, waterways, plains, or other topographies. Communications in such cases may be difficult, expensive, and time-delayed, and in some cases impossible. There are advantages in being able to permit real-time and / or continuous communications with remote or far-away areas, for example, the ability the monitor and respond quickly to changing conditions, weather, situations, and emergencies, and the ability to maximize efficiency of operations. Remote train lines are no exception to these issues. Many remote train lines are cut off from traditional communication lines, resulting in situations where a passing train may be out of communications with headquarters for an extended period of time. In such a situation, anymajor (such as an emergency, for example a derailment) or even minor (such as a rail blockage or weather obstruction) occurrence may be difficult to detect, trouble-shoot, and satisfactorily address, potentially leading to line delays, unexpected expenses, and in extreme cases catastrophic losses due to cascading events.

[0036] A mesh network may be used for local and in some cases external communications in an area. A mesh topology may be a network setup. Each network device may be interconnected with one another. Even in the case where a connection fails, a mesh network may permit transmissions to be distributed. A mesh network (or simply meshnet) may comprise a local area network topology in which the infrastructure nodes (i.e., bridges, switches, and other infrastructure devices) connect directly, dynamically, and non-hierarchically to as many other nodes as possible and cooperate with one another to efficiently route data to and from clients. A wireless network may comprise an infrastructure made up of devices that are wirelessly coupled to each other. The devices may help forwarding packets for one another so that the network can cover a larger area without the user needing to set up a dedicated infrastructure or wired connection for each device. The lack of dependency on one node, and decentralized decision-making and node cooperation may allow for every node to potentially participate in the relay of information. Mesh networks may dynamically self-organize and self-configure, which can reduce installation overhead. The ability to self-configure enables dynamic distribution of workloads, particularly in the event a few nodes should fail. This in turn contributes to fault-tolerance and reduced maintenance costs. Mesh topology may be contrasted with conventional star / tree local network topologies in which bridges / switches are directly linked to only a small subset of other bridges / switches, and the links between these infrastructure neighbours are hierarchical. While star-and-tree topologies are very well established, highly standardized and vendor-neutral, vendors of mesh network devices have not yet all agreed on common standards, and interoperability between devices from different vendors is not yet assured.

[0037] One or more wayside facilities may be used with a mesh network, such as a Hot Box Detector (HBD) bungalow. Such facilities may provide gateway points to distribute data remotely to a server back-office. Such facilities may incorporate long-range transmission equipment, such as long-range modems, including 4G LTE modems, or satellite capabilities. More than one type of long-range transmission equipment may be incorporated to provide redundant communication paths to the server.

[0038] Mesh networks may relay messages using either a flooding technique or a routing technique which makes them different from non-mesh networks. With routing, the message may be propagated along a path by hopping from node to node until it reaches its destination. To ensure that all its paths are available, the network may allow for continuous connections and must reconfigure itself around broken paths, using self-healing algorithms such as Shortest Path Bridging and TRILL (Transparent Interconnection of Lots of Links). Self-healing allows a routingbased network to operate when a node breaks down or when a connection becomes unreliable. As a result, the network may be quite reliable, as there is often more than one path between a source and a destination in the network. Although mostly used in wireless situations, this concept can also apply to wired networks and to software interaction.

[0039] Mesh networks may have a variety of features and parts. A mesh network whose nodes are all connected to each other is a fully connected network. Fully connected wired networks have the advantages of security and reliability - problems in a cable affect only the two nodes attached to it. However, in such networks, the number of cables, and therefore the cost, goes up rapidly as the number of nodes increases. By contrast, in a partiallyconnected network, nodes may be connected to neighbouring or adjacent nodes, creating a subnet, but without direction connections with all other nodes in the network. Mesh networks may contain one or more of gateways or border routers, mesh devices or routers, and leaf devices. A border router, sometimes interchangeably referred to as a gateway router, may be the interface to the outside world, and may connect the wireless mesh network to a building Ethernet, or connecting to a cloud service via Internet, for instance via LTE (Long Term Evolution) or 3G. A mesh device may be a device that helps build up the actual mesh, and may act to forward, or route, data to and from other nodes hence being the mesh backbone. Leaf devices may be devices that are part of the mesh network, but do not help creating the infrastructure. Such may not forward traffic on behalf of others but may communicate via the network. Leaf devices are often devices that needs to conserve energy due to constraints on battery. In a wired mesh, shortest path bridging and TRILL, may each allow Ethernet switches to be connected in a mesh topology, and allow for all paths to be active. IP (Internet Protocol) routing may support multiple paths from source to destination.

[0040] A wireless mesh network (WMN) may be a network made up of radio nodes organized in a mesh topology. Such may also be a form of wireless ad hoc network. In telecommunications networks, a node (Latin: nodus, ‘knot’) may be either a redistribution point or a communication endpoint. The definition of a node depends on the network and protocol layer referred to. A physical network node may be an electronic device that is attached to a network, and is capable of creating, receiving, or transmitting information over a communication channel. A passive distribution point such as a distribution frame or patch panel may consequently not be a node. Radio communication may be the technology of signaling and communicating using radio waves. Radio waves include electromagnetic waves of frequency between 30 hertz (Hz) and 300 gigahertz (GHz). Such may be generated by an electronic device called a transmitter connected to an antenna which radiates the waves and received by another antenna connected to a radio receiver. Radio is very widely used in modem technology, in radio communication, radar, radio navigation, remote control, remote sensing, and other applications. A wireless mesh may be self-healing if it can automatically repair itself when the environment changes. It could for instance be that a link between two nodes that worked perfectly fine an hour ago is now blocked by a bookshelf, a truck, or a steel door. The network will then automatically, without the involvement of any user, change its topology to be able to route the traffic a different path.

[0041] Mesh networks may have various advantages and disadvantages. In a full mesh or fully connected mesh network, the nodes within the network may be connected with every other. For example, if there are n number of nodes during a network, each node will have an n-1 number of connections. A full mesh provides an excellent deal of redundancy, but because it is prohibitively expensive to implement, it’s usually reserved for network backbones. The partial mesh may be more practical as compared to the full mesh. In a partially connected mesh, all the nodes aren’t necessary to be connected with one another during a network. Peripheral networks are connected using partial mesh and work with a full-mesh backbone in tandem. Advantages of Mesh Topology may include one or more of: a) failure during a single device won’t break the network, b) there may be no traffic problem as there is a dedicated point to point links for every computer, c) fault identification is straightforward, d) the topology provides multiple paths to succeed in the destination and tons of redundancy, e) such provides high privacy and security, f) data transmission is more consistent because failure doesn’t disrupt its processes, g) adding new devices won’t disrupt data transmissions, h) the topology has robust features to beat any situation, and i) a mesh doesn’t have a centralized authority.Disadvantages of Mesh Topology include one or more of: a) may be costly as compared to the opposite networktopologies i.e. star, bus, point to point topology, b) installation is extremely difficult in the mesh, c) power requirement is higher as all the nodes will need to remain active all the time and share the load, d), complex processes are needed, and complex algorithms, e) the cost to implement mesh is above other selections, f) there is a high risk of redundant connections, g) each node requires a further utility cost to think about, and h) maintenance needs are challenging with a mesh.

[0042] Embodiments of this disclosure may include protocols that aim to create a reliable, reactive, and power efficient system for a low power wireless communication network in remote area where infrastructure for constant power and hi-speed internet is not available. Such protocols may aim to solve the challenges of wireless communication over a long linear infrastructure such as rail tracks, mining conveyor, pipeline where power options are limited, and where mainstream communication modes, such as cellular network, satellite network are spotty, expensive, or non-existent. In some cases, the systems disclosed herein may be used on any communication applications, whether residential, industrial, remote, urban or other variety. Such protocols may dynamically allocate higher energy communications to adapt to the local limitations created by the variation of natural sources of power such as solar and wind over long distance, linear infrastructure and selects the most power smart communication mode in real time. Such protocols may be used as an application layer for any low power wireless mesh network. A core component of some of these protocols is the dynamic assignment of a border router role based on the availability of energy, wireless wide-area network (WWAN) link quality, low power wireless mesh network link quality and microcontroller unit (MCU) utilization of the node for any smart nodes on the network.

[0043] A communications system, according to the present disclosure, may provide a cost-effective overlay system for remote communications and data transfer. The system may be used to transfer data from a locomotive or other railway or trackside equipment to data processing centres for analytics. The system may also be used for point- to-train communications in remote areas where other forms of wireless communications, such as cellular networks, are unavailable.

[0044] Referring to Fig. 9, a communications system 60 is illustrated comprising a plurality of nodes 62. The plurality of nodes 62 may be distributed within a geographical area 66, such as a transport corridor as shown. Each of the plurality of nodes 62 may be in communication with one or more of the other of the plurality of nodes 62 to collectively form at least a partially connected wireless mesh network 70. Referring to Fig. 9, each of the plurality of nodes 62 may be configured for short-range, and in some cases long-range communications. Each node 62 may comprise one or more of a node processor 102, a battery, and a connection, such as connection, to a power source, such as a solar panel (not shown). Each node 62 may comprise a short-range transceiver 104 for short-range communications between nodes 62 of the plurality of nodes 62. Each node 62 may comprise a long-range transceiver 106 for long-range communication with a long-range network 68, which may interface with and / or include the internet 100. Each of the plurality of nodes 62 may be configured to dynamically assign and use one or more of the plurality of nodes 62 as a border router 72. The plurality of nodes 62 may be, in use, used to communicate with each other in a partially connected mesh network 70 and dynamically assign one or more of the plurality of nodes 62 as a border router 72. The border router 72 may be used to relay communications from the plurality of nodes 62 throughthe long-range network 68. The assignment of border routers 72 may be flexible and dynamic, and in some cases node specific.

[0045] Referring to Fig. 9, the geographical area 66 may comprise a remote area. In some cases, the area 66 comprises a transport corridor 64, such as a railway track for a train 22. The plurality of nodes 62 may be configured to maintain the train 22 on the railway track in continuous communication with the long-range wireless network 68. In some cases, the nodes 62 may cooperate with sensor units 28 to receive data about the train 22 and / or track, and relay that data off-site through the network 68, for example to an end user of a back-end system (control unit 78). The geographical area 66 may comprise a remote geographical area with portions or the entirety of which being out of contact with a cellular network (such as network 68) other than via one or more border routers 72, for example if one or more gaps 96 in network coverage are present along the corridor 64 or area 66.

[0046] Referring to Fig. 9, the plurality of nodes 62 may comprise one or more gateway routers 74. The one or more gateway routers 74 may be characterized by superiority, for example in long-distance communications or power supply, relative to the other of the plurality of nodes 62. Superiority may be measured on the basis of one or more of battery performance, long-range link quality, power source capacity, power source permanence, and the existence if any of a wired connection to the long-range network 68. Mere preferable location may be sufficient to assign a particular node 62 as a gateway router 74. In a first mode, each of the plurality of nodes 62 may assign and use one or more gateway routers 74 as a border router 72 in preference to the other of the plurality of nodes 62. The first mode may be the default mode or may be a mode selected when all other variables are neutral as between candidate border routers 72 or in favor of the gateway router 74. In a second mode, each of the plurality of nodes 62 or one or more of them may assign and use another (not the gateway router 74) of the plurality of nodes 62 as a border router 72, for example when a ranking of the gateway router 74 drops below a predetermined threshold or below a ranking of the another of the plurality of nodes 62, such as nodes 62’ and 62’”” in the example of Fig. 1. It should be understood that in the drawings and description, the use of apostrophe suffixes (such as ‘or ”) on a reference character refers to a type of the part identified by the reference character, and references may be made in the description or drawings with or without the suffix and refer to the same part in the drawings or description, without or with the suffix, respectively. In some cases, a gateway router 74 may have the same infrastructure and parts as a regular or average node 62, and in some cases as all of the nodes 62.

[0047] Referring to Fig. 9, each of the plurality of nodes 62 may be configured to dynamically assign a border router 72 for the node 62. Assignment of a border router 72 may be based on rankings dynamically assigned to each of the plurality of nodes 62. A ranking of each node 62 may be based on one or more of the following performance characteristics: battery capacity, battery performance, available charging capacity, long-range link quality, microcontroller unit (MCU) utilization, number of other nodes in communication range with the node, number of other nearby candidate border routers, distance to the nearest suitable candidate border router, long-range network load, and short-range bandwidth. The ranking of each node 62 may be based on at least battery capacity, available charging capacity, and long-range link quality, in some cases.

[0048] Referring to Fig. 9, each node 62 may be configured to broadcast status information about itself to other nodes 62, to be used to assign appropriate border routers 72 for each node 62. In some cases, each of theplurality of nodes 62 are configured to broadcast to other of the plurality of nodes 62 on performance characteristics of the node 62. Broadcasting by a node 62 may be carried out based on a suitable timing, such as at periodic intervals, according to a schedule, and / or as a result of a change in performance characteristics. In some cases, new rankings may be broadcasted at time intervals of an hour or less, for example 30 minutes or less, in some cases 10 minutes or less, or at other suitable intervals longer or shorter than the aforementioned examples. Each of the plurality of nodes 62 may be configured to self-rank and broadcast a self-ranking for the node 62 to other of the plurality of nodes 62. By self-ranking itself and broadcasting a self-ranking for the node 62 to other of the plurality of nodes 62, a node 62 may be efficiently providing sufficient information to nodes 62 to reduce analytical burdens on those other nodes 62 when self-assigning border routers 72. Each of the plurality of nodes 62 may be configured to dynamically assign and use one or more of the plurality of nodes 62 as a border router 72 based on the rankings provided by each of the plurality of nodes 62. Assignment may be based on both the self-rankings received from other nodes 62 and a short- range link quality between the respective nodes 62. Short-range link quality may be a function of a variety of suitable factors, such as historical message success rate, latency, network adjacency, and time since last advertisement. Once assignment of an appropriate border router 72 is achieved, a message may be transmitted by a node 62 to the border router 72. The message is then relayed by the border router 72 out of the network 70, for example to a long-range network 68 and beyond.

[0049] Referring to Fig. 9, the nodes 62 may have suitable transceivers to enable short and long-range communications. Each node 62 may have a long-range transceiver 106. Transceiver 106 may comprise one or more of: a cellular transceiver, a satellite transceiver, a broadcast radio transceiver, and a microwave transceiver. The long- range wireless network 68 may comprise one or more of the internet 100, a cellular network 68, and a satellite network 98. The short-range transceiver 104 may comprise one or more of a low frequency radio transceiver, an infrared transceiver, a Bluetooth transceiver, a Wi-Fi transceiver, and a mesh network transceiver. In some cases, transceivers 104 and 106 are provided on the same part, for example as part of a multi-function transceiver system. The short-range transceiver 104 may communicate at suitable frequencies, for example as low frequency radio transceiver. The short-range transceiver 104 may communicate at suitable frequencies and may comprise a 2.4 GHz or 915 MHz wireless transceiver. Other ranges may be used. Smart node transceivers 104 and 106 may include long- range high-power transceivers (i.e., WWAN) and short rang low power transceiver (i.e., for a low power wireless mesh network).

[0050] Fig. 9 illustrates a conceptual layout of the relationship and basic parts of the system 60. Each node 62 may be associated with one or more sensor units 28. Each node 62 may be connected to receive data from a sensing device or device, such as a sensor unit 28. Sensor units 28 may be configured to sense and provide information on one or more of motion, temperature, vibration, tilt, rail or corridor integrity, seismic activity, humidity, water levels, weather, flooding, proximity, obstacle, wildlife, sound and visual elements. The sensor units 28 may be located on or adjacent the corridor 64, and / or on the train 22. In the example of Fig. 3, each node 62 may receive data from sensor units 28 and may either communicate the data off-network via a respective on-board transceiver 106, or route the data to another suitable border router, such as gateway router 74 in the example shown, passing through one or more other nodes 62 in the process. The data may be ultimately sent via the network 68, for example the internet 100, to a suitable user, such as a back-end client, for example a back-end server 80. In the example shown, a back-endserver 82 receives data, and a front-end server 84 relays, and / or analyzes and provides information via a front-end system accessible by an end user.

[0051] Referring to Fig. 9, the system 60 may communicate with a suitable back-end system / control unit 78. A back-end system / control unit 78 may comprise a server processor, such as server 80, 82, and / or 84, connected to receive and transmit communications from and to, respectively, the plurality of nodes 62 via the long-range network 68. A storage device 88 may be used, such as a computer readable medium for storing algorithms and data. The storage device 88 may store information in one or more databases. One or more displays 86 (and input devices, such as keyboards, mice, and others) may be connected in the system / control unit 78 to permit a developer or server operator to access and manipulate the back-end system / control unit 78. The system / control unit 78 may be located a substantial distance away from the network 70, for example 50, 100 or more miles away. In some cases, the system / control unit 78 may be located in a city or town, or in a different province, state, or country than the network 70. In some cases, a node 62 can hop 30-100 nodes. If a node 62 is used on every mile, hopefully in a 100-mile segment you would have at least one node with connectivity. If no connectivity is established over a suitable time, the node 62 may queue data for trying at another time, but such may be an indication that the user should build up the network further. Data may be dumped from memory after a certain period of time, for example 4 hours. Assuming that a train will move into the area, however, and data storage could be transported to a mobile station (i.e., the train), the mobile station could become a carrier for data to get to an area with long range connectivity. The server 80 may be connected to serve and relay communications information from the plurality of nodes 62 to third parties via the long-range network 68. In one case, the front-end server 84 may deliver usable forms and content, for example by JavaScript processes or other processes suitable for display in a web browser, accessible by the client or user, for example on a display 94 of a computer 92 of a client or user, or on a mobile phone 90 of the client or user. Thus, a user of the system / control unit 78 may use a phone 90 or computer 92 to monitor in real-time or near real-time the status of the network 70 and any relevant aspects within the network 70, for example the progress and integrity of track or train 22.

[0052] As shown in Fig. 9, the communications system 60 may have a plurality of track-side nodes 62, which are installed alongside a railway line or track. A plurality of gateway routers 74 may be installed periodically adjacent to the railway line or corridor 64, to facilitate communication between one or more adjacent nodes 62 and a long-range network 68, such as a cellular network. Alternatively, the long- range network 68 may be another form of long-distance wireless communication, such as a satellite network, or it may be a long distance wired communication network, such as a fibre-optic cable or copper wire network. When a locomotive (train 22) travels over a section of railway line in a remote region without access to a long-range network 68, such as in a gap in cellular coverage, the train 22 may wirelessly communicate with one or more nearby nodes 62. The node 62 relays the information to adjacent nodes 62 until the signal reaches a node 62 in range of the nearest gateway router 74. The signal is then transmitted to a data processing centre system / control unit 78, via the long-range network 68 where it can be analyzed and a response sent, if necessary.

[0053] Each node 62 may have a processor, a power source, and a wireless transceiver, which are mounted within a housing. Preferably, the power source is a battery charge by a solar cell mounted on or near the node 62. Other power sources may be used, where available, such as a wired power supply, a wind power generator, or otherform of available harvested energy. The processor may control the electrical components of the node 62 and may be mounted inside the housing. Preferably, the processor is a microcontroller unit (MCU), or central processing unit (CPU) installed on a printed circuit board (PCB) with the other electronic components of the node 62. The nodes 62 may also have a display mounted within the housing. Preferably, the display is a touch display, such as an OLED (Organic Light-emitting Diode) touch panel, to facilitate easy user interface for diagnostics, maintenance, or other such tasks. Other electrical components which may be mounted within the housing of each node include: memory, storage, and peripheral connections, such as USB or serial ports, or other types of data ports.

[0054] The wireless transceiver may use any suitable type of wireless communication, such as satellite communication, broadcast radio, microwave communication, cellular network, infrared communication, Bluetooth, Wi-Fi, mesh network, or other suitable type of wireless communication. Preferably, each node 62 may be capable of communicating with a cellular network for long distance communication (i.e., with the long-range network 68), via an IEEE ™ (Institute of Electrical and Electronics Engineers) transceiver, and with a mesh network for short distance communication with adjacent nodes 62, via a low-frequency radio transceiver.

[0055] The use of a low-frequency radio transceiver for node-to-node communications provides a longer range for node-to-node communications and lower power consumption compared to many other types of wireless communication. Using this type of wireless transceiver to communicate node-to-node within the mesh network also permits fewer nodes to be used to cover long stretches of railway lines in remote regions where there are gaps in cellular coverage. The long-range of such a wireless transceiver also permits nodes 62 to communicate with nonadj acent nodes 62 and, thereby, provides some redundancy to the mesh network, in the event of a failure or communication error with one node 62. In the event of such a failure, the signal could “skip” the affected node 62 and thereby maintain communication between the other nodes 62 and the long- range network 68.

[0056] The gateway routers 74 are optional and, preferably, each node 62 has multiple wireless transceivers or a single wireless transceiver capable of multiple types of wireless communication to enable the nodes 62 to also perform the function of the gateway routers 74. Where gateway routers 74 are used, they are preferably installed on or with existing track-side communication equipment to facilitate communication over existing communication infrastructure. For example, a gateway router 74 could be installed where existing track-side equipment is already connected to a local power grid and wired (or wireless) communication network, to facilitate communication between nearby nodes 62 and the available connection to a long-range network 68, such as a fibre-optic network or cellular network.

[0057] Where gateway routers 74 are not used, each node 62 may be able to communicate via both short and long-distance wireless communication. Alternatively, only selected nodes 62, which are installed within range of a long-range network 68, may be configured for both short and long-distance wireless communication, to reduce the cost of nodes 62 located in gaps in cellular coverage. As a result, nodes 62 can operate both as a “local” node, communicating via the mesh network, and an “uplink” node, communicating via the long-range network 68. Nodes 62 may be assigned as either local nodes or uplink nodes (border routers) by the remote data processing centre system / control unit 78 or automatically. Local nodes may leave their long-distance wireless transceiver, such as a cellular modem, powered off to save power.

[0058] Preferably, uplink nodes are assigned automatically and dynamically by software running on each node 62 communicating over the mesh network. The software analyzes each node’s 62 performance, based on a number of performance factors, to identify the best candidates for uplink nodes. The performance factors may include: battery performance, number of other nodes in communication range, cellular signal strength, number of other nearby candidate uplink nodes, distance to the nearest suitable candidate uplink node, or other relevant factors. For example, a node with consistently stronger cellular signal strength than other nodes would be a better candidate for an uplink node. Similarly, a node in range of more than one other node would be a better candidate, as would a node with consistently better battery performance (due to better positioning of its solar panel, etc.). The various performance factors may be weighted to produce an overall value that permits identification of the best candidate nodes.

[0059] The system 60 may also periodically re-evaluate candidate nodes 62, and re-assign nodes 62 as local nodes or uplink nodes, when appropriate. For example, due to the additional power required for communication with the long-range network 68, an uplink node may drain its battery below a certain threshold, triggering another nearby local node to be re-assigned as an uplink node, at least until the former node can recharge its battery and resume service as an uplink node. In this way the system is able to dynamically assign uplink nodes to maintain optimal power consumption and a desired number of uplink nodes for optimal communications.

[0060] In operation, the system 60 is installed trackside along a railway line to facilitate communication with a locomotive or other railway or track-side equipment. In one exemplary embodiment, the system may be used to facilitate communications for a train protection system, such as a positive train control (PTC) system. In such a system, information on track conditions ahead of a locomotive is sent to a control unit on the locomotive, which adjusts the speed of the train accordingly. Interruptions of communication with the locomotive’s control system, such as occurs in remote areas with gaps in cellular coverage can compromise safety or train operations. For example, if track conditions ahead of a locomotive change after it enters a region of railway line in an area with a gap in cellular coverage, it may be too late for the control unit to safely slow the train by the time the locomotive exits the area and re-establishes communication. The communications system of the present disclosure provides an alternative method of communicating with the control unit on the locomotive in such remote areas to enable uninterrupted communication of information on track conditions.

[0061] If information on track conditions changes, such as in the event an emergency condition is detected, while the locomotive is out of communication contact with a long-range network the emergency condition would be transmitted to the nearest gateway router or uplink node, via the long-range network. The emergency condition would then be transmitted node-to-node, via the mesh network, until it reaches a node within communication range of the locomotive. The emergency condition is then transmitted, via the same type of wireless communication used between nodes, preferably a low-frequency radio transceiver, to the control system on the locomotive. The control system may then take appropriate action, such as applying the brakes, in response to the emergency condition, while the train is still within the region of railway line located in a gap in cellular coverage before the locomotive is able to re-establish a direct connection with the long-range network. This provides additional time for the control system on the locomotive to respond to the emergency condition, thereby improving train safety and reducing the risk of derailments, collisions, or other serious incidents.

[0062] In addition to being used to facilitate communications with a long-range network in remote regions, the system may also be used to broadcast information locally within the mesh network. For example, if the system is connected with track-side sensors or other safety equipment to transmit that information to a data processing centre, the nodes or gateway routers may also have software to analyze the data from track-side sensors to determine if the data indicates an emergency condition. The system may then broadcast the emergency condition for a set number of node-to-node relays, from the node(s) detecting the condition, which may vary depending on the nature of the emergency condition. For example, a condition requiring a train to slow before passing through a certain section of track may be broadcast through the mesh network for a distance equal to the distance required for a train to slow from its normal travelling speed on that section of track to the reduced speed required by the track condition. This local signal may be broadcast, regardless of whether the signal is being transmitted to the data processing centre, via a gateway router or uplink node. This permits the control system on a locomotive to receive the information on the emergency condition even in the event of a communication failure between the system and the data processing centre.

[0063] Referring to Figs. 10-13, example plots are illustrated of voltage waveforms (an example of voltage waveform data) and processed voltage waveform data (also an example of waveform data) from plural sensor units. Fig. 10 illustrates, for a single sensor unit, how readings might differ under different situations. The units on the y-axis of the plot may be relative numerical representations of voltage detected by the sensor unit over time while the sensor unit carries out a stimulus electrical voltage. In the example shown, the stimulus electrical voltage begins around about 11 ms, and continues until about 23 ms. In the example shown, the numerical representation unit is referred to as an ADC count, which is an analog reading that is a numerical representation of the voltage level, as a direct representation of voltage but in units that do not correspond with actual voltage, as such units may be unit factored up or down for the purpose of analysis and emphasis. Each overall waveform shows a time-based plot of average readings during a signal pulse of the sensor or another sensor. The relative voltage measured is not a constant over this time period, due to the impedance of the railroad track. Impedance refers to the resistance and reactance, which is the property of an electrical component to oppose the change in current. With opposition to flow being the highest at the beginning of the pulse, and thus the initial pulse has a higher voltage at the start of the pulse. As current starts to flow in the rail, the resistance to flow decreases, as a function of the impedance, leading to voltage decay as illustrated. As shown, the voltage waveform 150 is highest in magnitude under an open rail (broken rail event) event, due to the presence of a break adjacent the sensor unit lowering the capacity of the railroad to flow current past the break, increasing the impedance of the system and the corresponding voltage detected. As well, the voltage waveform 154 is the lowest under a train present situation, as the axles of the train act as shunts, greatly reducing the impedance of the system, and the recorded voltage as well. The voltage waveform 152 under the normal state is in between the waveforms 150 and 154, corresponding to normal operation without breaks or trains.

[0064] In order to convert the waveforms into data that can be monitored to illustrate the changing status and environment of a railroad over time, the waveforms from each sensor unit may be processed into processed waveform data. Referring to Figs. 11-13, plots of processed waveform data are illustrated for a three-sensor unit system monitoring a section of railroad track. In order to create the plots illustrated, for each waveform detected by a respective sensor unit during a pulse (electrical stimulus voltage), whether originating from that sensor unit or anothersensor unit, the waveform is converted into a representative value. In the examples shown, averages are used, being the average numerical representation of voltage over the course of an electrical stimulus voltage. An average is used as an approximation to reflect proportional changes, and is not a perfect quantification of relative voltage detected, however, same may be sufficient to compare data over time in a section of track to draw certain conclusions of state. Other calculation methods may be used to generate the processed waveform data. Thus, each point in the plots of Figs. 11-13, may be calculated using the average of the ADC (Analog to Digital Converter) count between start and end of pulse, which can be visualized by taking the average of the ADC count between 13-22 ms in the data of Fig. 10 for one such point.

[0065] The plots shown in Figs. 11-13 include respective lines 158, 160, and 162 corresponding to averaged waveform data of sensor units arbitrarily labeled 24.1 0, 25.2 0, and 26.2 0, respectively, and that are located in the same order in spaced relation along the section of track. In each of the three figures, the sensor unit with the highest relative data is the sensor unit that is providing the stimulus electrical voltage, while itself and the other sensor units are detecting the effects of that voltage. In the figures, an event is detected around between 8:00 and 10:00 between sensor units 24.1 0 and 25.2 0. Upon initial detection of the relatively increased waveform data at around 8:00 in the various Figures, further readings may be required to identify the event, which could be from a broken rail or loose joint event. A loose joint could occur anywhere on the track and can modify a good electrical connection to no electrical connection, and can vary in between. A broken rail event may be reflected by the same initial voltage waveform pattern as a loose joint event, with the difference between the two typically being the fact that a loose joint event may self-correct over time (and particularly across temperature transitions that might occur during the day), whereas a broken rail event may remain indefinitely in the absence of a repair. In the example shown, a repair crew may have been initiated shortly after 8:00 to check the status of the line. Upon checking the line between sensor units 24.1 0 and 25.2 0, a broken rail was discovered and repaired by 10:00. Thereafter, the voltage waveform data returned to pre-break level as shown. In other cases, the data may reveal a loose joint event. The control unit may be configured to determine whether an event corresponds to a loose joint event by a suitable method, such as by detecting, after initial detection of the broken rail event, for example at around 8:00, relatively less to no, attenuation if detectable, and elevation, of voltage waveform data, for example decaying immediately or after a given amount of time. Thus, in the example shown, the waveform data remains relatively constant between 8:00-10:00, and then the relative elevation and attenuation drop off in the recorded data. In other words, the sensor units after 10:00 may no longer detect a broken rail event, in the event that no repair has occurred on the section of track. In such a case it is believed that a shift in data from broken to continuous rail in the example may be caused by temperature changes occurring to the track, for example occurring as the day progresses due to the action of the sun, leading to different stresses on the rail. For example, in the morning or evening, the rail may experience the most significant temperature transitions during the day, for example by relative cooling to go under tension (shortening) at dusk, or by relative heating to go under compression (lengthening) at dawn), making it more likely that a loose joint event would be detected at such times. In some cases, loose joint events are most likely to occur just at dawn, where the railroad is likely to be at its lowest relative temperature during a daily cycle, all other factors being equal. Similarly, during daylight hours, and moreso on sunny days during daylight hours, the relatively higher ambient temperature from the sun will warm the rail, putting the rail under compression (lengthening), reducing the negative effects of the loosejoint, and in some cases, eliminating the detection of the loose joint event altogether (as shown). In some cases, a loose joint is most likely to occur at the neutral temperature. The neutral temperature is the temperature at which the rail assumes an orientation where it is neither under tension nor compression. The control unit may be configured to determine whether the broken rail event corresponds to a loose joint event by detecting a relative lessened elevation and attenuation occurring during daylight hours, for example the hours between sunup and sundown, or during a shorter time period between sunup and sundown. In the example shown, the charts also depict a number of train present events, for example at around 4:00, 14:00, 15:30, 20:00, and 22:00.

[0066] A DC-based system may stimulate the rail with relatively low voltage or extra low voltage (such as 5 V, although in some cases voltages of less than or equal to 35 V may be used, such as voltages less than or equal to 15 V) direct current pulses at predetermined intervals, in order to observe any variation in the signal response measured by sensor units. DC sensor units may be configured to operate individually or in groups. Sensor units may interact with adjacent sensor units in the same group. If one individual or standalone sensor unit is used for a track segment, the operator may install shunts at predetermined distances on both sides of the sensor unit. For a group of sensor units, the operator may install shunts at a predetermined distance from the outermost sensor units in the group. The sensor units may be spaced from one another along the track segment with sufficient coverage (via proximity) by individual sensor units such that data from multiple units may be collected and used by the back-end server to identify and corroborate a break in the track segment even when one or more sensor units adjacent the break fail to function properly. Such coverage may be obtained by arranging the sensor units in close enough proximity such that each sensor unit’s detection range (coverage) overlaps each other's coverage areas to obtain this benefit from group system installation. Overlapping coverage may refer to the fact that more than one sensor unit may be close enough to each section of the track segment such that a break can be identified even if the sensor unit or units directly adjacent the break are non-functioning. In the examples shown, sensor units may be located close enough to one another such that sensor units two or more sensor units away will detect changes 41 and 43 sufficient to trigger the control unit to initiate a response indicative of a possible event. When multiple units are used, they may provide overlapping coverage to create redundancy within the system, where one or more units may fail without the system losing the ability to detect important events during operation. In some cases, a response may be triggered by a 7% increase from rolling average in transmission mode. In some cases, a response may be triggered by a 20% increase / decrease in receiving mode. A suitable threshold employed provides a confidence level that can be applied to detected changes 41 and 43 to provide the operator with confidence that an event has in fact occurred. Rolling averages may be continually updated, for example to respond to environmental changes. If a track section experiences a weather change, the data collected may gradually update the rolling average of each sensor unit accordingly, to avoid false positives and to ensure accurate detection of events in changing weather conditions. Levels can change from environmental factors like rain, so rolling averages may be modified in such cases. Data that is within the validity range is considered to be relevant and is used to update the rolling average.

[0067] The system may be configured to monitor a rail section past a switch point or on the opposite sides of a switch that separates the spur track from the main track, or between two switches that divide or separate the siding track from the main track. Switches can present as either a short circuit or as an open circuit independent of the position of the points. The electromechanical connections in a switch can be electrically opened or closed by themechanical forces distributed throughout the switch during reversal or normalization. The electrical properties can also be affected similarly by the forces exerted on the electromechanical connections during a train pass. This can result in unstable electrical properties of the monitored circuit. Where the section of railroad track comprises a switch, the switch may have shunts on either side of the switch and forms a blind zone in the section. In some cases, all switch joints intended to pass current through the switch are bonded. In some cases, insulated joints are installed or present on the switch to prevent the switch from shunting the rails in operation. In some cases, bond wiring may be provided with sufficient tension such that a broken rail between the bonded section would be expected to sever the bond wire. Switch components may be upgraded with insulated components to prevent same from acting as a shunt. To prevent spur tracks and siding tracks from serving as an alternate pathway for DC signals rather than using the main track, insulated joints may be installed in such cases on both sides of the siding tracks and on the spur tracks that pass the switch. To maintain strength, durability, and longevity, specific track equipment and components, including switch frogs, guard rails, heel blocks, and switch points, may be constructed from high-quality steel. The conductive property of such equipment provides an alternative path for DC signals. The gradual variation in system signal values caused by a change in the conductivity of the ballast and ties at low temperatures and during precipitation may suggest that environmental factors must also be periodically analyzed in order to determine the threshold values and design a system that can compensate for those factors in order to enhance the performance of the system.

[0068] The present disclosure has been described and illustrated with reference to an exemplary embodiment, however, it will be understood by those skilled in the art that various changes may be made, and equivalents may be substituted for elements thereof without departing from the scope of the invention as set out herein. Therefore, it is intended that the invention is not limited to the particular embodiments disclosed herein.

[0069] Referring to Figs. 14-19, the control units or unit 78 may be configured to reduce alternating current (AC) interference present on the railroad track. AC present in power lines can induce an AC current in adjacent railroad tracks due to electromagnetic induction. As AC flows through the power lines, it generates a time-varying magnetic field around the conductors. According to Faraday's Law of Electromagnetic Induction, this changing magnetic field induces an electromotive force (EMF) in nearby conductive materials, such as railroad tracks. The magnitude of the induced current depends on factors such as the strength of the magnetic field, the distance between the power lines and the tracks, the frequency of the AC supply, and the electrical properties of the tracks. This phenomenon can cause unwanted currents, potentially interfering with track signaling systems and posing safety risks in railway operations. AC interference may also be present on the railroad track electrical signal as a result of other devices configured to apply an AC stimulus to the railroad track. The AC-induced interference may be identified in voltage waveform data measured by the plurality of sensor units 28. As AC -induced interference, the interference signal will have a defined wavelength based on frequency, typically 60 Hz in North American power lines (60 Hz is shown in the example, but this frequency may be different than 60 Hz, for example in different jurisdictions), that can be identified in the waveform data when no stimulus is being applied to the railroad track. Fig. 14 illustrates waveform data 200 where AC -induced interference is present, and no electrical stimulus is being applied. Fig. 15 illustrates waveform data 202 in an idealized case without AC-induced interference, with the sensor unit 28 detecting one or more pulses 204, 206 (two pulses are shown in the example) fired as the electrical stimulus on the railroad track. Figs. 16 and 17 illustrate the effect of AC -induced interference on an ideal signal, with the magnitude of thepulses 204 and 206 increased and diminished, respectively. The plot of Fig. 17 illustrates that, depending on the timing of the electrical stimulus, the AC-induced interference in voltage waveform data 208 from a sensor unit adjacent a power line may enhance the pulse signal, or diminish it, leading to unreliable results and reducing the ability of the system to detect events. It should be noted that, in the example of Fig. 17, reverse voltage circuit protection (such as via a diode) has been incorporated in the operation of the sensor unit 28, thus preventing negative voltages from operating in the circuit below a predetermined negative voltage threshold.

[0070] Referring to Figs. 18-19, the control unit 78 may be configured to reduce AC interference from the voltage waveform data 208, for example by shifting and overlaying the voltage waveform data on itself. Shifting may be carried out by duplicating the voltage waveform data (creating data 208’) by an interval of (n + ‘A )*period of the AC-induced interference, where n is an integer, and the period is a time interval equal to the time the AC -induced interference completes one period, i.e. a full cycle or wavelength. The duplicated waveform data 208’ is then overlaid (added to) the voltage waveform data 208, i.e. added onto itself. The shift and overlay may be carried out by transposing and adding individual sets of numerical representations of voltage in a time-series data set from the sensor unit. Shift and overlay may alone be effective where a sensor unit 28 does not employ reverse circuitry protection, however, where such is present, the user may wish to involve a further step of using two or more pulses 204, 206 spaced by an interval equal to the interval selected for the shift, i.e. (n+l / 2)*period as the electrical stimulus. Each pulse may have a suitable length of time, such as between 5 to 50 ms or other intervals. Fig. 19 illustrates processed voltage waveform data 210, which has been derived by overlaying and adding voltage waveform data on itself by an interval of 4.5 periods in the example. The use of two pulses 204, 206 results in three peaks 212, 214 and 216 in the processed voltage waveform data 210, with the middle peak 214 typically being the largest. The control unit 78 may identify one or more of the peaks 212, 214, and 216, and use the largest or the average peak as the representative data to use in the further analysis for rail events. In some cases of data, a receiving mode sensor unit may be unable to detect two pulses, for example if one of the pulses has a strong enough diminishment from the AC interference so to not be detectable. In some of these cases, the control unit may respond in a variety of ways, such as by electing not to shift and overlay, or by requesting that the cycle be repeated with an electrical stimulus of higher power. More than two pulses may be used. The pulses may all be separated by the interval or an integer multiple of the interval. The selection of a representative signal may be done after shifting, for example the shift may be carried out and the control may attempt to select the largest signal above a threshold baseline value, or may average the peaks (signals above threshold) that are located. Table 1 below illustrates examples for a variety of processed and raw voltage waveform data from the above-recited methodology.

[0071] Table 1 : comparison of pulse data in raw and processed waveform dataPulse No Noise Pulse Raw Received AC noise shift and Dual Offset Level Level overlay method Pulse Overlay1 100 % 189 % 161 %2 100 % 22 % 106 %3 100 % 89 % 112 %

[0072] Referring to Figs. 20-24, in some cases there may be multiple AC interference signals on the railroad line, and such may be noise cancelled to increase signal to noise as before. A railroad track may have multiple AC interference at different respective frequencies. For example, at a train crossing, there may be present both a 60 Hz AC-induced interference from an adjacent power line, and a different frequency of AC interference, such as an 86.7 Hz frequency AC signal in the example present as a result of a crossing warning system, or other systems and devices for other purposes. Figs. 20-22 illustrate the effects of dual frequency AC interference, via a waveform data 220 that is the combination of pulses 204 and 206, and waveform data 200. In such a case, noise cancellation may be carried out to reduce both sets of interference. As shown in Figs. 23 and 24, the control unit may be configured to reduce the AC interference from the voltage waveform data 220 by shifting, and overlaying the voltage waveform data 220’ on itself, by an interval that is selected to be equal to ( + ‘A )*periodi = (nm+ A^periodm, where n is an integer, 1 to m denote respective AC interferences of different frequencies, and period is the time of one wavelength of the respective AC interferences*. For the example charts m (nm) was chosen as being 1 integer value higher than m (where ni= 4) for the purposes of keeping the data charts compact and minimizing the time interval between pulses. This results in an example second noise waveform frequency of 86.7 Hz. In practice the actual periodicity of the interfering waveforms would be used to find integer values of ni, m through nmthat satisfy the equation. The resultant processed voltage waveform 224 may have more easily identified peaks at 212, 214, and 216 as before. The control unit may identify the peaks by identifying a threshold signal strength, and averaging any peak that exceeds that threshold. For example, if two peaks are located above a threshold signal, the average of both might be taken. If one peak is located, that peak may be taken as the value.

[0073] *The interval relationship for multiple frequencies may be written, for any number of plural frequencies, as being equal to each of:(m + A )*periodi(n2 + A )*period2(nm+1 / 2)*periodm,

[0074] In the claims, the word “comprising” is used in its inclusive sense and does not exclude other elements being present. The indefinite articles “a” and “an” before a claim feature do not exclude more than one of the feature being present. Each one of the individual features described here may be used in one or more embodiments and is not, by virtue only of being described here, to be construed as essential to all embodiments as defined by the claims.

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A railroad track sensor system comprising: a plurality of sensor units distributed at various respective locations along a section of railroad track that is bounded on opposed ends by rail-to-rail shunts, the plurality of sensor units configured to cycle through a rotation where each sensor unit is set to operate, one at a time, in a sending mode where the sensor unit produces a stimulus electrical voltage across the parallel rails, while the plurality of sensor units are set to simultaneously operate in a receiving mode where the plurality of sensor units each measure a resultant voltage across the parallel rails of the railroad track; and a control unit configured to compare voltage waveform data, from the resultant voltages measured by the plurality of sensor units during the rotation, with calibrated voltage waveform data to determine whether the section of the railroad is in a normal state, a broken rail state, or a train present state.2 The railroad track sensor system of claim 1 in which the control unit is configured to determine an event location, being a length of the section of railroad track between two or more sensor units, of the plurality of sensor units, or between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, where one or more of a broken rail event, a train present event, or a loose joint event has occurred.3 The railroad track sensor system of claim 2 in which the control unit is configured to determine an event location where a broken rail event has occurred, in use, between sensor units, of the plurality of sensor units, by detecting: elevated voltage waveform data at one or more sensor units, of the plurality of sensor units, on a first side of the event location, and attenuated voltage waveform data at one or more sensor units, of the plurality of sensor units, on a second side of the event location, when the one or more sensor units on the first side are in the sending mode; and attenuated voltage waveform data at the one or more sensor units on the first side of the event location, and elevated voltage waveform data at the one or more sensor units on the second side of the event location, when the one or more sensor units on the second side are in the sending mode.4 The railroad track sensor system of claim 3 in which the control unit is configured to determine the event location where the broken rail event has occurred, in use, between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, by detecting elevated voltage waveform data of the plurality of sensor units, with a relative elevation of the voltage waveform data decreasing, in send mode and receive mode, with increasing distance from the one of the opposed ends.5 The railroad track sensor system of any one of claim 3 - 4 in which the control unit is configured to determine that the broken rail event corresponds to a loose joint event by detecting, after initial detection of the broken rail event, relatively less elevation and, if previously detected, attenuation, of voltage waveform data.6 The railroad track system of claim 5 in which the control unit is configured to determine that the broken rail event corresponds to the loose joint event by detecting, after initial detection of the broken rail event, relatively less elevation and, if previously detected, attenuation, of voltage waveform data during at least neutral zone temperature transitions.

7. The railroad track sensor system of any one of claim 3 - 6 in which the control unit is configured to determine an event location where a train present event has occurred, in use, at a sensor unit, of the plurality of sensor units, or between adjacent sensor units, of the plurality of sensor units, by detecting a relatively largest attenuation in voltage waveform data of the sensor unit or adjacent sensor units, when in send mode and receive mode.

8. The railroad track sensor system of claim 7 in which the control unit is configured to determine an event location where a train present event has occurred, in use, at a sensor unit, of the plurality of sensor units, or between adjacent sensor units, of the plurality of sensor units, by detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from the sensor unit or adjacent sensor units in send mode and receive mode.9 The railroad track sensor system of any one of claim 1 - 8 in which the control unit is configured to determine the existence of an event, comprising one or more of a broken rail event, a train present event, or a loose joint event, that has occurred, by detecting one or more of an attenuation or elevation of voltage waveform data at one or more sensor units, of the plurality of sensor units, beyond a predetermined threshold change.10 The railroad track sensor system of claim 9 in which the predetermined threshold change corresponds to a 5% elevation or greater for determining the existence of a broken rail event or a loose joint event detected by a sensor unit of the plurality of sensor units, in send mode.11 The railroad track sensor system of any one of claim 9 - 10 in which the predetermined threshold change corresponds to a 10% elevation or greater, or a 20% attenuation or greater, for determining the existence of the event by a sensor unit, of the plurality of sensor units, in receive mode.12 The railroad track sensor system of any one of claim 1 - 11 in which the control unit is configured to determine calibrated voltage waveform data for one or more of the sensor units, of the plurality of sensor units, based on rolling averages when in the normal state.13 The railroad track sensor system of any one of claim 1 - 12 in which the section of railroad track comprises a switch, and one or more of: the switch has shunts on either side of the switch and forms a blind zone in the section; all switch joints intended to pass current through the switch are bonded; or insulated joints are installed or present on the switch to prevent the switch from shunting the rails in operation.14 The railroad track sensor system of any one of claim 1 - 13 in which the rail-to-rail shunts comprise one or more of un-isolated switches, or specifically placed boundary shunts.15 The railroad track sensor system of any one of claim 1 - 14 in which the control unit comprises a plurality of control units.16 The railroad track sensor system of any one of claim 1 - 15 in which the control unit is connected to send instructions and receive data wirelessly from the plurality of sensor units.17 The railroad track sensor system of any one of claim 1 - 16 in which the control unit comprises a remote server that is accessible by the plurality of sensor units via the internet.18 The railroad track sensor system of any one of claim 1 - 17 in which the plurality of sensor units are distributed along the railroad at even intervals.

19. The railroad track sensor system of any one of claim 1 - 18 in which the plurality of sensor units comprises three or more sensor units.

20. The railroad track sensor system of any one of claim 1 - 19 in which the plurality of sensor units are configured to produce the stimulus electrical voltage by direct current (DC).

21. The railroad track sensor system of any one of claim 1 - 20 in which the control unit is configured to reduce alternating current (AC) interference from voltage waveform data measured by the plurality of sensor units.

22. The railroad track sensor system of claim 21 in which the control unit is configured to reduce AC interference from the voltage waveform data by shifting, and overlaying the voltage waveform data on itself, by an interval of (n + A period of the AC-induced interference, where n is an integer, and the period is the time of one wavelength of the AC interference.23 The railroad track sensor system of claim 22 in which: the AC interference is present at two or more frequencies; and the control unit is configured to reduce the AC interference from the voltage waveform data by shifting, and overlaying the voltage waveform data on itself, by an interval that is selected to be equal to, for each AC interference, (nm+ / )*periodm= (nm+ ‘A)* periodm, where n is an integer, 1 to m denote respective AC interferences of different frequencies, and period is the time of one wavelength of the respective AC interferences.24 The railroad track sensor system of any one of claim 22 - 23 in which, the plurality of sensor units are configured to, in the sending mode, produce a stimulus as a plurality of pulses spaced by the interval.25 A method of monitoring a section of a railroad track that is bounded on opposed ends by rail-to-rail shunts, with a plurality of sensor units distributed at various respective locations along the section of railroad track, the method comprising: cycling the plurality of sensor units through a rotation where each sensor unit operates, one at a time, in a sending mode where the sensor unit produces a stimulus electrical voltage across the parallel rails, while the plurality of sensor units are simultaneously operated in a receiving mode where the plurality of sensor units each measure a resultant voltage across the parallel rails of the railroad track; and determining whether the section of the railroad is in a normal state, a broken rail state, or a train present state, by comparing voltage waveform data, from the resultant voltages measured by the plurality of sensor units during the rotation, with calibrated voltage waveform data.26 The method of claim 25 in which determining further comprises determining an event location, being a length of the section of railroad track between two or more sensor units, of the plurality of sensor units, or between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, where a broken rail event, a train present event, or a loose joint event has occurred.27 The method of claim 26 in which determining further comprises determining the event location where a broken rail event has occurred, in use, between sensor units, of the plurality of sensor units, by detecting: elevated voltage waveform data at one or more of the sensor units, of the plurality of sensor units, on a first side of the event location, and attenuated voltage waveform data at one or more of the sensor units, of the plurality of sensor units, on a second side of the event location, when the one or more sensor units on the first side are in the sending mode; andattenuated voltage waveform data at the one or more sensor units on the first side of the event location, and elevated voltage waveform data at the one or more sensor units on the second side of the event location, when the one or more sensor units on the second side are in the sending mode.

28. The method of claim 26 in which determining comprises determining the event location where the broken rail event has occurred, in use, between one of the opposed ends of the section of railroad track and a sensor unit, of the plurality of sensor units, by detecting elevated voltage waveform data of the plurality of sensor units, with a relative elevation of the voltage waveform data decreasing, in send mode and receive mode, with increasing distance from the one of the opposed ends.

29. The method of any one of claim 27 - 28 in which determining comprises determining that the broken rail event corresponds to a loose joint event by detecting, after initial detection of the broken rail event, relatively less elevation, and, if previously detected, attenuation, of voltage waveform data.

30. The method of any one of claim 26 - 29 in which determining comprises determining the event location where a train present event has occurred, in use, at a sensor unit, of the plurality of sensor units, or between adjacent sensor units, of the plurality of sensor units, by detecting a relatively largest attenuation in voltage waveform data of the sensor unit or adjacent sensor units, when in send mode and receive mode.

31. The method of claim 30 in which determining comprises determining the event location where a train present event has occurred, in use, between adjacent sensor units, of the plurality of sensor units, by detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from the adjacent sensor units in send mode and receive mode.

32. The method of claim 30 in which determining comprises determining the event location where a train present event has occurred, in use, at an adjacent sensor unit by: detecting a relative local maximum attenuation of voltage waveform data at the adjacent sensor unit in send mode; and detecting attenuated voltage waveform data of the plurality of sensor units, with a relative attenuation of the voltage waveform data decreasing with increasing distance from sensor units adjacent to the adjacent sensor unit.

33. The method of any one of claim 25 - 32 in which determining comprises processing voltage waveform data to reduce alternating current (AC)-induced interference.

Images