WO2025048740A1

WO2025048740A1 - A wireless mesh network to receive and/or send data from/to beneath the rubble

Info

Publication number: WO2025048740A1
Application number: PCT/TR2024/050451
Authority: WO
Inventors: Alphan Sahin; Huseyin Arslan
Original assignee: Istanbul Medipol Universitesi
Current assignee: Istanbul Medipol Universitesi
Priority date: 2023-08-28
Filing date: 2024-05-06
Publication date: 2025-03-06
Anticipated expiration: 2026-02-28

Abstract

The invention relates to a method and associated system for locating and contacting people trapped under rubble as soon as possible, using mesh networks of self-improving nodes with detection capabilities operating under and above the rubble.

Description

A WIRELESS MESH NETWORK TO RECEIVE AND/OR SEND DATA FROM/TO BENEATH THE RUBBLE

Technical Field

State of the Art

Although someone who is under the rubble is conscious and has a mobile phone, it is very difficult to establish a connection with cellular networks. The main reason for this is that the collapsed concrete, brick, and iron piles weaken the radio signals (from outside to inside or from inside to outside) significantly (approximately 30-60 dB [1]). Another reason is that the external cellular networks are probably damaged or inaccessible by the earthquake.

If a person under the rubble does not have any communication devices, unfortunately, contacting them depends largely on the search and rescue teams and their equipment. In the literature, the technologies used to find those under the rubble can be made by means of heat, sound, or CO2 measurements [2], Although there are systems operating on electromagnetic or acoustic waves with radar principles (for example, NASA FINDER, which aims to detect heartbeats from 5-10 meters with signals sent in the 3 GHz band), in general, since the signals sent from these devices are significantly weakened, they are not suitable for life detection in the depths of the rubble [3], [4],

In the literature, LoRa nodes are commonly used over the LoRaWAN (Long Range Wide Area Network) standard. LoRaWAN is a low-power wide-area network specification based on star topology. With this topology, nodes communicate bi-directionally wirelessly over unlicensed bands (863-870 MHz and 2.4 GHz in Turkey) with gateways. Gateways transfer data from nodes to central servers, and users can interact with the data by accessing central servers. Since it allows bi-directional unlicensed communication, allows nodes to run on batteries for up to 10 years, and its relatively low costs per node ($5-$50), it can now be used in a wide range of applications such as environmental monitoring, forest fire monitoring, lighting systems, automation, underground communication, and irrigation systems. However, due to the star topology, the coverage area of LoRaWAN is limited to the coverage area of the gateways. In addition, gateways can be damaged due to earthquakes. In a similar way, gateways may not be able to perform their duties due to the fact that the Internet is not working. Therefore, a typical LoRaWAN network will not be sufficient in the face of a disaster such as an earthquake.

In this research, instead of star topology, how mesh topology should work in earthquake conditions for LoRa will be investigated. With mesh topology, communication will take place by transferring over nodes instead of gateways. Since LoRaWAN is an open code system, it allows such a change. However, as of now, there is no such update. In the literature, there are preliminary research studies on mesh topology. For example, in the study of Meshtastic [5], it was shown that communication between two nodes for a very large distance of 166 km can be made over mesh topology. However, this study is based on flooding and is not designed for a disaster such as an earthquake.

In the literature, it has been shown that an approximate location information of a node in LoRaWAN can be extracted [6], However, position prediction algorithms based on the received signal strength indicator do not produce accurate results and can create an estimation error of around 50-100 meters. Therefore, finding the positions of the nodes under the rubble will not work. Another problem is that the positions of the nodes change randomly due to the earthquake and there is no self-improvement due to the lack of necessary reference nodes. In this study, in order to solve this problem, nodes will estimate their distances to each other with the principle of round-trip time-of-flight (RTToF). With this method, how long it takes for a signal to be calculated from one node to another and the distance between the two nodes can be estimated by multiplying it by the speed of light. It is possible with Semtech LoRa SX1280 chips operating at RTToF 2.4 GHz frequency [7], It has been shown in the literature that position information can be estimated with an error of 1 meter with the presence of regular reference nodes (anchor nodes) [8], [9],

All the problems mentioned above have made it necessary to make an innovation in the relevant field as a result. Objects and Brief Description of the Invention

The main object of the invention is to provide communication with the area under the rubble.

The object of the invention is to determine the locations and/or environment data of the persons and/or objects under the said rubble.

The object of the invention is to reduce conflicts between communication nodes induced by the rubble, which negatively affect communication.

Brief Description of the Invention

In order to eliminate the above-mentioned problems, a system using communication nodes placed before the earthquake and communication nodes placed on or around the rubble after the earthquake and a method suitable for this system are used. Here, the system autonomously establishes the network in a way to reduce conflicts between nodes with the algorithm that can find periodic role changes, self-correct slots, and timings, and it is ensured that nodes find their own positions over the progressively established network.

With the establishment of the network autonomously to reduce inter-node conflicts, the mesh network can be organized to reduce self-interference without any class difference between nodes.

It provides the location of active nodes scattered under the rubble by finding their own position over the progressively established network. Here, an autonomously established mesh network is used.

Here, it is preferably recommended that message transmissions be made according to the position information of the nodes in the network in order to deliver messages generated from the nodes under the rubble (for example, messages such as help messages, messages for remote life or object detection, or alarm situations such as heat or gas leakage) to the reference nodes. In other words, it is recommended that transfers over nodes are not made randomly but based on their own location in the network from nodes under the rubble to reference nodes. In this way, messages will be transmitted in a shorter time and more nodes will consume less power

Definitions of Figures Describing the Invention

The figures and related descriptions used to better explain the device developed by this invention are as follows.

Figure 1. A schematic view of the system subject to the invention showing the installation in an area containing the building and its surroundings

Figure 2. A schematic view of the system subject to the invention on the collapsed building and in an area containing its surroundings

Figure 3. Distribution of LoRa nodes after rubble and representation of nodes that can communicate with each other (N_anchor=25, N_node=50, dth=25 m). The filled circles show the reference nodes, and the hollow circles show the nodes under the rubble.

Figure 4. A time diagram showing the periodic operation of any node as an initiator or responder.

Figure 4a. A time diagram showing that the slots are not aligned with each other and cause interference because the nodes activate at different times.

Figure 5. Common and hidden nodes of nth and mth nodes.

Figure 6. Selection of sample graph and most suitable slot (N_siot=10,N_node=16).

Figure 7. Illustrations showing the nodes finding each other by the progressive location method.

Figure 8. Illustrations showing the working principle of ranging with RTToF.

Figure 9a. For RTToF, the ranging request sent from the initiator and responder nodes and the change in time of the ranging response packets (SF=5,BW=800 kHz)

Figure 9b. Spectrograms of ranging request and ranging response packets sent from initiator and responder nodes for RTToF (SF=5,BW=800 kHz)

Figure 9c. For RTToF, the ranging request sent from the initiator and responder nodes and the change in time of the ranging response packets (SF=6,BW=800 kHz).

Figure 9d. For RTToF, the ranging request sent from the initiator and responder nodes and the spectrograms of the ranging response packets (SF=6,BW=800 kHz).

Figure 10. Illustrations showing the proposed protocol for ranging with RTToF. Figure Ila. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=50, dth=15 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure 11b. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=50, dth=20 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure 11c. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=50, dth=25 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure lid. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=100, dth=15 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure lie. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=100, dth=20 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure Ilf. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=100, dth=25 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure 11g. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=150, dth=15 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure llh. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=150, dth=20 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure Hi. For Nnode and dth values, the graph showing the change in time of the probability of conflicts of at least one node (N_node=150, dth=25 m, p=0.25, Tinitiator=20 ms, T_period=l s, NanchorE{25,36,49,64,81,100}).

Figure 12a. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (N_node=50, dth=l 5 m, p=0.25, Tinitiator=20 ms, T_period=l S, NanchorE{25,36,49,64,81,100}).

Figure 12b. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (N_node=50, dth=20 m, p=0.25, Tinitiator=20 ms, T_period=l S, NanchorE{25,36,49,64,81,100}). Figure 12c. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (N_node=50, dth=25 m, p=0.25, Tinitiator=20 ms, T_Period=l S, NanchorG{25,36,49,64,81,100}).

Figure 12d. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (Nnode=100, dth=15 m, p=0.25, Tinitiator⁼20 HIS, Tperiod⁼l S, NanchorG {25,36,49,64,81, 100}).

Figure 12e. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (Nnode=100, dth=20 m, p=0.25, Tinitiator⁼20 HIS, Tperiod⁼l S, NanchorG {25,36,49,64,81, 100}).

Figure 12f. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (Nnode=100, dth=25 m, p=0.25, Tinitiator⁼20 HIS, Tperiod⁼l S, NanchorG {25,36,49,64,81, 100}).

Figure 12g. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (Nnode=150, dth=15 m, p=0.25, Tinitiator⁼20 HIS, Tperiod⁼l S, NanchorG {25,36,49,64,81, 100}).

Figure 12h. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (Nnode=150, dth=20 m, p=0.25, Tinitiator⁼20 HIS, Tperiod⁼l S, NanchorG {25,36,49,64,81, 100}).

Figure 12i. In case of conflicts for Nnode and dth values, the graph showing the change in time of how many nodes observed interference on average (Nnode=150, dth=25 m, p=0.25, Tinitiator⁼20 HIS, Tperiod⁼l S, NanchorG {25,36,49,64,81, 100}).

Figure 13a. On average, the graph showing the change of the position information of the node under the rubble according to the steps (Nnode=50, NanchorG {25,36,49,64,81, 100} and dth=15).

Figure 13b. On average, the graph showing the change of the position information of the node under the rubble according to the steps (Nnode=50, NanchorG {25,36,49,64,81, 100} and dth=20).

Figure 13c. On average, the graph showing the change of the position information of the node under the rubble according to the steps (Nnode=50, NanchorG {25,36,49,64,81, 100} and dth=25).

Figure 13d. On average, the graph showing the change of the position information of the node under the rubble according to the steps (Nnode=100, NanchorG {25,36,49,64,81,100} and dth=15). Figure 13e. On average, the graph showing the change of the position information of the node under the rubble according to the steps (N_node=100, NanchorG {25,36,49,64,81,100} and dth=20).

Figure 13f. On average, the graph showing the change of the position information of the node under the rubble according to the steps (N_node=100, NanchorG {25,36,49,64,81,100} and dth=25).

Figure 13g. On average, the graph showing the change of the position information of the node under the rubble according to the steps (N_node=150, NanchorG {25,36,49,64,81,100} and dth=15).

Figure 13h. On average, the graph showing the change of the position information of the node under the rubble according to the steps (N_node=150, NanchorG {25,36,49,64,81,100} and dth=20).

Figure 13i. On average, the graph showing the change of the position information of the node under the rubble according to the steps (N_node=150, NanchorG {25,36,49,64,81,100} and dth=25).

Figure 14a. On average, the graph showing the analysis of the position information of the node under the rubble (Nnode=50, NanchorG {25,36,49,64,81,100} and dthG { 10,15,20,25}).

Figure 14b. On average, the graph showing the analysis of the position information of the node under the rubble (Nnode= 100, NanchorG {25,36,49,64,81,100} and dthG { 10,15,20,25}).

Figure 14c. On average, the graph showing the analysis of the position information of the node under the rubble (Nnode=l 50, NanchorG {25,36,49,64,81,100} and dthG { 10,15,20,25}).

Definitions of Components/Pieces/Parts of the Invention

In order to better explain the device developed by this invention, the parts and pieces in the figures are numbered and the corresponding numbers are given below.

N. Communication node

RN. Reference node

Detailed Description of the Invention The subject matter of the invention relates to a method and a system for locating and contacting people trapped under rubble as soon as possible, using mesh networks of selfimproving nodes with detection capabilities operating under and above the rubble.

Referring to Figure 1 and 2, the present system comprises multiple communication nodes (N). Said communication nodes (N) are preferably long-range known as "LoRa" and operate in the Long-Range Wide Area Network standard known as "LoRaWAN".

In order to carry out the existing system, at least a part of the existing communication nodes (N) must be placed on the building before the building is turned into rubble and a part of the communication nodes (N) must be positioned outside the building. Communication nodes (N) placed outside the building, there is a difference in terms of work before or after the building is turned into rubble. Nodes placed outside the building can also be expressed as reference nodes (RN) and preferably reference nodes (RN) and communication nodes (N) are electronic devices with the same feature and therefore reference nodes (RN) are also expressed as communication nodes (N) in this document.

The number of communication nodes placed in the building is expressed as A_node. It is also called A_node binding node. Here, the identity information (ID) of the nth communication node (N) is expressed by When the building collapses, the

positions of the pre-positioned communication nodes are expressed as L_x, L_y and L_z, and L_x, L_y and L_z show the coordinates in a three-dimensional area. When the building collapses, communication nodes (N) are activated. The position information of the nth communication node in this case in the Cartesian system is shown by the vector.

Following the demolition of the building, for example after the earthquake, the lV_anchor reference node (RN) or nodes of the search and rescue teams are placed on the rubble to form a grid in the x and y planes. In this study, the reference nodes (RN) know their position information precisely and are placed on the x and y axes, preferably at equal intervals (For example, lV_anchor = b², b E IL is provided). The position of the reference nodes on the z axis is T_z. Preferably, all communication nodes (N) operate as time-domain duplexing (TDD) and on the same channel and with the same physical properties (for example, the same spreading factor and bandwidth). If the distance between the nth and mth communication nodes (N) is less than the d_th threshold distance, that is,

, these two nodes can communicate with each other. For example, if

and the rubble size is the location of 75 nodes and the nodes that can

communicate with each other are exemplified in Hata! Ba$vuru kaynagi bulunamadi.

In this study, the identity information of the neighboring nodes of the nth communication node is expressed by the

set, wherein

After the earthquake, first of all, the mesh network between the nodes should be established autonomously. The objects at this stage are to be able to find the identity information of the nodes neighboring the nth LoRa node, for example N_n, Yn E {1,2, — , N_noAe + !V_anchor], and to minimize interference situations that may occur due to the use of the same band for later communication situations. The main problem is that there is no coordinator to organize the network. Therefore, there is a need for an autonomous solution that can organize the mesh network by working by nodes on their own.

In the present invention, it is first proposed to periodically change the method to reduce interference in the mesh network for autonomous installation. Here, for the installation of the mesh network autonomously, it is first recommended that all nodes act as "initiator" and "responder" (role) periodically in two modes during T_period. During this time, any node can only actively access the channel during time T_initiator and only works as a receiver during the rest of the time. It is possible for nodes to behave according to different modes as well as "initiator" and "responder" modes. For example, nodes may contain "detection" and/or "calculation" modes. However, initiator and responder modes are required for the basis of the invention. If the duration of these modes is determined in advance, it is thought that the relevant values are included in the translation time calculation.

In addition, it is thought that a node can skip the initiator role in this study. The node assumes the initiator role with the predetermined p E [0,1] probability for this behavior (or will skip the initiator role with the 1 — p probability). If p = 0.5 taken as an example, nodes will on average skip the initiator role by half and continue to behave as responders. Skipping the initiator role is important and necessary for resolving inter-node interference and the slotbased solution proposed below.

In the present method, the

rate is taken as an integer. Therefore, the time zone is divided into A_slot slots. At this point, it should be noted that any node can also act as a receiver during T_initiator. This is especially necessary for the ranging mentioned in the next section. The periodic behavior of any node is shown in Hata! Ba$vuru kaynagi bulunamadi..

In the event of an earthquake, all nodes will activate at different times. To indicate this fundamental difference, the local time information of the nth node is expressed by t_n (for example, this information may be the time since the microprocessor first started operating). In addition, the slot index, in which n assumes the role of the initiator of the node, is denoted by s_n E {1,2, ... , lV_slot}. The moment when the nth node can work by taking the initiator role using local time information is expressed as the moment when the following equation is achieved

(these moments are indicated by an arrow in Hata! Ba$vuru kaynagi bulunamadi.). In Equation (1), the value is defined as the translation time. The reason for including this

parameter in (1) is that even if the communication nodes change mode periodically as in (N) Hata! Ba$vuru kaynagi bulunamadi., they do not act synchronously at first because they are independent of each other. For this reason, the slots of the nodes may not be aligned in real time and the packets may not reach the surrounding nodes due to potential interference. The problem of not aligning the slots for three nodes is shown in Hata! Ba$vuru kaynagi bulunamadi..

By translating with the At_n parameter in time, it is aimed that the nodes move synchronously with each other, and the slots are aligned. Therefore, in order to establish the network autonomously, it is necessary to be able to learn the nth node At_n and a_n values, Vn, by itself. The optimization problem for slot selection is defined as follows when all nodes slots are aligned (i.e. when ideal ] values are calculated):

As shown in Figure 5, any two nodes can have common nodes and hidden nodes. Here, the function gives how many own and neighboring mth nodes use the

same slot as their neighbors (except for itself) when using the nth node and the slot. In

Figure 6, the most suitable slot selection for A and (assumed to be

^anchor ⁼ 0) for ^a grid graph is exemplified. It is assumed that a node in the group here is within communication distance with the nodes next to it. As can be seen, although the number of slots is smaller than the number of nodes, is always zero for any two

nodes and the best solution can be obtained. As seen here, slots have spatial reuse. For example, the n = 2th and n = 15th nodes use the 10th slot.

Preferably, in the present method, the optimization problem is solved without any coordinator, i.e. by all nodes using the same algorithm and communicating only with their neighbors, in order to autonomously reduce any nth node s_n and 4t_n values in the mesh network conflicts,

1) It broadcasts the identity and slot information of its neighbors detected so far with the probability of p when it skips the initiator state.

2) If the nth node receives announcement information from the mth node: a. If the slot used by itself, for example, s^ \ 1) conflicts with one of the slots specified in the announcement packet (that is, the slot of the node from which the packet came, for example, the slots used by the neighbors of the s£\ and mth node identified so far), or 2) if it is not itself listed as a neighbor, it will randomly select a slot that will not conflict. If there is no conflict, it continues to use the same slot. b. The slot of the node where the packet came from saves s® and identity information. c. It selects the translation parameter as

according to the announced slot of the incoming packet and the time of its slot.

Herein, the flow in the n'th node, which finds the s_n and At_n parameters autonomous to reduce interference, is described as follows:

Definitions:

• Announcement address: This is a number that all virtually defined nodes listen to. In this study, the announcement address was accepted as ADR=0x00 (as HEX).

• Announcement command: This is a number that indicates that an announcement is made, which all virtually defined nodes know in advance. In this study, the announcement command was accepted as CMD=0x00 (as HEX). In general, a different number can be selected.

• set with the identity information of the neighbor of the

nth node detected at the moment of t.

• set containing the slot information of the D_n neighbor of the

nth node detected at the moment of t.

Algorithm outputs:

•

Setup phase:

• is randomly selected from the set.

•

Loop phase:

• If (1) is provided: o if provided, (i.e. if the nth node has decided to work as the

initiator):

■ With the announcement address and announcement command, they will broadcast their own ID information and the number of the slot they are currently using, for example, a_n and

the ID information of their neighbors detected up to them, for example, and the slot

information they use, for example

• Else, (i.e. if the nth node is in responder mode): o If a packet is received and the ADR area of the incoming packet is 0x00 and the CMD area is 0x00 (in this case, the incoming packet comes from the a_m node and the slot it uses is s_m and the information of its neighbors is shown with

and S

■ If a

■ Else:

■ If < is an empty set:

• s used by its neighbors):

is not an empty set, select a

new random value from this set to

•

■ Else:

• (if slot s_n is used by its neighbors and by

the neighbors of node m) or a (if it is not itself

specified as the neighbor of node m)

•

not an empty set, select a new random value from this set to

■ (Set the translation parameter to

the announced slot of the incoming packet and the time of your slot)

At this point, it should be emphasized that:

In a system where the translation time methods are embedded (such as a microprocessor), the relevant state-machine can be realized by translating the transition timing from one state to another.

In addition to slots, the announcement packet can also carry extra information about the number of transfers, position information, or nodes. For example:

1. A node that broadcasts the announcement packet: a. It can indicate in the announcement packet whether it is a reference node. b. It can indicate its transfer number in the announcement packet. c. It can indicate in the announcement packet whether it knows its place. i. If it knows its place, it can indicate its place in the announcement packet.

2. Its neighbors: a. It can indicate in the announcement packet whether its neighbor is a reference node. b. It can indicate the transfer numbers in its neighbor's announcement packet. c. It can indicate in the announcement packet whether its neighbor knows its place. If its neighbors' places are known, it can indicate their places in the announcement packet.

Apart from the responder and initiator modes, it is predicted that the nodes may have different modes such as detection or calculation, if the duration of these modes is determined in advance and the relevant values can be included in the translation time.

In the present invention, there is a position of a communication node (N) positioned outside the rubble, capable of communicating with a communication node (N) in the rubble and remaining in the rubble with at least one communication node (N) whose position is known, and the position of another communication node (N) under the rubble with the communication node (N) in its position is progressively located.

Communication nodes (N), which can be accessed to reference nodes (RN) under the rubble at the beginning, will predict their own places by using the position information of reference nodes (RN). Then, the communication nodes (N) that find their place will be a reference to other communication nodes (N). With this chain structure, it is recommended to determine the place of all nodes progressively. After the nodes detect their own places, they will share the information whether there is life in the vicinity of the sensors on them with their own location information with the search and rescue teams.

In order for a node to determine its place in 3D space, that node must have four neighbors who know its place at least. In this study, it is shown by d® that the location of the nth node is not detected in the t step. If d® = 1, the nth node shows that it can locate itself in the tth step (that is, it has at least four neighbors who know its own location), and if it is d® = 0, the node cannot locate itself. In this case, the determination of a node's own location is progressively expressed on the graph as follows:

In this study, it is accepted that the parameter of the reference nodes is 1 and

all other nodes under the rubble are

Figure 7 provides examples for the proposed progressive positioning method (lV_node = 50,

^anchor = 25, d_th = 25 m). It shows the initial state and none of the nodes under the rubble know their place, for example = O.In the first step (t = 1), the nodes that can

detect their own places are shown as filled circles. For example, node 7 under the rubble can find its place because it can communicate with reference nodes 66, 67, 71 and 72, for example At this stage, 11 nodes under the rubble can find their place. It was shown that 41

nodes were located at the time of t = 2. The number of nodes that can find their place at moment t = 3 is 49. Again, as shown in Figure 7, only node 20 cannot find its place. The reason for this is that there are not four neighbors which know its place.

The more nodes there are, the more likely that node is to find its own location. However, this also increases the likelihood of interference. For this reason, A_node and lV_anchor parameters should be selected for a certain number of slots, and both communication and location measurements should be selected in the most appropriate way.

In this study, positioning measurements were made with Semtech SX1280 chips. The distance between this chip and the two nodes is measured by RTToF. RTToF sends a ranging request packet by specifying the address of a certain node as a node initiator. Let's show the duration of this packet in discrete time by A_req length. Let's suppose the signal sent from the initiator node reaches the responder node after T_p seconds. At this stage, the initiator node begins to count the ticks of its own clock at a certain frequency. Let's show this frequency as f_A. Another responder node at that address receiving the ranging request packet sends the ranging response packet after a certain time and reaches the initiator after T_p seconds. If the initiator node knows how long an array is used in the responder node, for example, N_res can calculate the T_p value as follows:

If /A ⁼ FA is provided and this value is assumed to be known, the distance between the two nodes can be found by multiplying by the speed of light. The operating principle of ranging with RTToF in Hata! Ba$vuru kaynagi bulunamadi. is given.

Smallest time interval in LoRa chip is given as follows:

In other words, the ideal values of fa and fa are 2¹² BW. Although this corresponds to a fairly high resolution (e.g., (speed of light)), it is

emphasized in [10] that this does not imply high accuracy. LoRa chips know their own N_res times and lV_measure = lV_{cnt A} — N_res gives them directly after measurement. In this case, the uncalibrated ranging can be calculated as follows:

Theoretically, the lowest possible value of the standard deviation for distance to LoRa chips can be found with the Cramer-Rao lower bound given below

Here, ^Vindicates how many measurement values are taken and c indicates the speed of light. As can be seen, the more measurements are taken, the higher the bandwidth is used, or the higher the dispersion parameter is selected (i.e., the longer a packet is sent), the less dispersed the measurement results will be.

In Hata! Ba$vuru kaynagi bulunamadi., the ranging request sent from the initiator and responder nodes for RTToF and the changes in time and frequency of the ranging response packets are given. In this measurement, the bandwidth was selected as 800 kHz, and the SF parameters were selected as 5 and 6. Signals were measured in the air with Adalm Pluto SDR. As can be seen, the packet time increases as the SF parameter increases. At the same time, the chirp symbols are clearly visible. At this point, it should be noted that the chirp time is defined as follows in LoRa chips:

Figure 9a-9d shows that the signal sent from the responder contains at least 16 chirp symbols.

In this study, if any node has at least four neighbors which know its place, it is recommended to measure RTToF with these neighbors. For this, if the nth node obtains the right to use its own slot, that is, when (1) is provided, it broadcasts the start-ranging packet by using the address of the relevant neighbor to measure the distance with one of the said neighbors. Selects the ranging address and command for this packet as follows:

• Ranging address: This is a virtually defined number that only the interested node listens to. In this study, the identity information of the node to be communicated was defined as the address, a_n, (as HEX).

• Ranging command: This is a number that all virtually defined nodes know in advance. In this study, the announcement command was accepted as CMD=0x01 (as HEX). In general, a different number can be selected.

As seen in Hata! Ba$vuru kaynagi bulunamadi., if node A indicates the address of node B, node A sends the start-ranging packet with node B. Node B changes its mode as the ranging responder after receiving the start-ranging packet and waits for the ranging request packet. Node A broadcasts the ranging request packet and node B receiving this packet broadcasts the ranging response packet. Then, node A records the range value. After a node measures the distance with at least four neighbors who know its own place, it finds its own location with the existing methods in the literature (for example, the least squares method) [11] [12] [13],

Finally, in the present method, data is sent outside the rubble using communication nodes (N). It is recommended that message transmissions be made according to the position information of the nodes in the network in order to deliver messages generated from the nodes under the rubble (for example, messages such as help or life detected messages) to the reference nodes. In other words, it is recommended that transfers over nodes are not made randomly but based on their own location in the network from nodes under the rubble to reference nodes. Two basic methods have been proposed for this.

In the first method, communication outside the rubble is carried out with a minimum number of communication nodes (N). Here, the information of how many steps any node can reach the reference nodes is used. The number of transfers for this method, for example, parameter h_n, is defined as:

For reference nodes,

As a transfer rule, the nth node under the rubble transfers its message to one of its neighbors with the lowest h_n value to deliver the message to any reference node. Therefore, the message is gradually approaching the reference node.

For this method to work:

1. A node that broadcasts the announcement packet specifies the value of its

parameter in the announcement packet. If the h_n parameter is not clear at the time, it indicates it.

2. A node receiving the announcement packet saves its neighbor's h_m parameter and updates its h_n parameter according to

With this method, messages can be received from a node that does not know the location.

In the second method, the h_n value indicating in which step the communication node (N) detects the position during the progressive position discovery is defined and the communication node (N) to transmit the message to the lowest -h_nvalued communication node (N) to which it can communicate, and this method is repeated until the message reaches outside the rubble.

It is progressively based on the node positioning method. With this method, if the nth node can find its position in the tth step during progressive node positioning, the number of transfers of the node to the reference nodes is defined as follows:

For reference nodes,

As a transfer rule, the tnh node under the rubble transfers the message to one of its neighbors with the lowest h_n value to deliver the message to any reference node. Therefore, the message is gradually approaching the reference node as in the first method.

According to the first method, at least four of the neighbors meet the following condition in accordance with the principle of progressive node positioning:

In other words, there are more alternatives.

For this method to work:

Progressively, if the nth node finds its position in the tth step during node positioning, it records the number of transfers of the node to the reference nodes as

A node that broadcasts the announcement packet specifies the value of its h_n parameter in the announcement packet. If the h_n parameter is not clear at the time, it indicates it.

1. A node receiving the announcement packet saves its neighbor's h_n parameter.

The disadvantage of this method is that the positions should be made for communication. If a node does not know its place, it does not transmit a message. However, since the location of the transferring node is known, it eliminates uncertainties. Since there are at least four alternative nodes, it is more durable than the first method.

In Table 1, the proposed packet structures are detailed in order to use the methods proposed in the previous sections.

Table 1. Definitions of announcement, start-ranging, and message packets.

Three basic packets are defined as shown in Hata! Ba$vuru kaynagi bulunamadi.

The announcement packet gives the information of a node, the information of its neighbors, identity details, position information, slot information, and whether it is a reference node. If the position information of a node is not known, the position information areas are sent.

The start-ranging packet specifies with which node the distance will be measured. The message packet sends the desired message to a node selected according to the h_n value. CMD=0x02 was selected for this packet. This number can preferably be selected as a different number.

The solutions proposed by the present invention were tested by simulating with MATLAB. The simulations were made assuming that and .

The message packet sends the desired message to a node selected according to the h_n value. CMD=0x02 was selected for this packet. In general, a different number can be selected.

In Figure 1 la-1 li, the change in time of the probability of conflicts of at least one node for and p = 0.25

values are given. As the number of neighboring nodes in the graph increases (that is, the dth> Miode and ^anchor values increase), the probability of conflicts increases. However, the algorithm proposed in the present invention for all scenarios reduces the conflicts over time. In other words, all nodes choose their own slots and translation parameters to reduce interference.

In case of conflicts in Figure 12a- 12i, the number of interference observed on average varied over time. The results conflicting with those in Figure 1 la-1 li, and the number of nodes exposed to interference on average decrease over time. The number of nodes affected increases in proportion to the A_anchor, A_node and d_th values. There are cases where the number of nodes affected increases with time when there is conflict between Hata! Ba$vuru kaynagi bulunamadi.d and Hata! Ba$vuru kaynagi bulunamadi.g. However, when viewed from Hata! Ba$vuru kaynagi bulunamadi.d and Hata! Ba$vuru kaynagi bulunamadi.g, the proposed algorithm reduces the probability of such a situation.

In Hata! Ba$vuru kaynagi bulunamadi., the change of the average number of nodes that can extract the position information for A_node E {50,100,150},

and

is given according to the steps, for example, As

expected, the more the number of nodes that can communicate with each other, the shorter the time the nodes can detect their own places. For example, the highest success rate can be achieved in 2-3 steps for d_th = 20 m, while this value increases to 5-6 steps for d_th = 15 m. While the d_th G {10,15,20,25} and iV_anchor = {25,36,49,64,81,100} values are given in Hata! Ba$vuru kaynagi bulunamadi., the analysis of the position information of the node under the rubble is given on average. If a node is adjacent to at least four positioned nodes, it can find its place. As can be seen from the results, the more intensely the nodes in the graph are connected to each other, the success rate of finding the location of any node increases. The intensity of the graph is directly proportional to the magnitudes of the d_th, A_node, and iV_anchor parameters. While parameter d_th allows more nodes to be connected to each other, A_node and ^anchor increase the likelihood that a node will be adjacent to more nodes.

In Table 2a, 2b and 2c, using the curves in Hata! Ba$vuru kaynagi bulunamadi.a-l li and Hata! Ba$vuru kaynagi bulunamadi., the conflict probability at the 50th second and the positioning success in the 25th step are given as different A_node, lV_anchor and d_th values. As can be seen, the positioning success and the reduction of the probability of conflicts are inversely proportional to each other. In Table 2, the probability of finding the node location is better than 90% and the probability of conflicts is less than 10%. For example, given d_th = 25 in Table 2a, the probability of conflicts A_node = 50 and iV_anchor = 25, 36, 49, 64 and 81 with A_node = 100 and iV_anchor = 25, 36 and 49 is less than 10%, and the probability of finding the node location for A_node = 50,100 ve 150 and iV_anchor = 25, 36, 49, 81 and 100 and is more than 90%. Similar values are given for d_th = 20 ve 15 in Tables 2b and 2c. As can be seen in Table 2a-2c, there are situations to consider both of these metrics. At this point, it should be emphasized that the most appropriate option among these solutions varies according to the situation. For example, more dense nodes will increase life sensing performance. However, this will also increase costs. The construction material used in the building will affect d_th. The d_th value will be lower for a steel building than for a reinforced concrete building. In such a case, more nodes need to be placed. In this case, more nodes are required for a steel building.

Table 2a. Selection of iV_node and iV_anchor parameters required to be able to locate the node while giving d_th = 25 and at the same time to keep interference low.

Table b3. Selection of iV_node and lV_anchor parameters required to be able to locate the node while giving d_th = 20 and at the same time to keep interference low.

Table 2c. Selection of lV_node and N_anchor parameters required to be able to locate the node while giving d_th = 15 and at the same time to keep interference low.

REFERENCES

[1] Holloway, C. , Koepke, G. , Camell, D. , Remley, C. , Schima, S. , McKinley, M. and Johnk, R. (2006), Propagation and Detection of Radio Signals Before, During, and After the Implosion of a Large Convention Center, Technical Note (NIST TN), National Institute of Standards and Technology, Gaithersburg, MD

[2] Zhang, Di, Salvatore Sessa, Ritaro Kasai, Sarah Cosentino, Cimarelli Giacomo, Yasuaki Mochida, Hiroya Yamada, Michele Guarnieri, and Atsuo Takanishi. 2018. "Evaluation of a Sensor System for Detecting Humans Trapped under Rubble: A Pilot Study" Sensors 18, no. 3: 852.

[3] NASA,

Last accessed: 01/31/2023

[4] V. Cable, J. Lux and S. Haque, "Target & propagation models for the FINDER radar," 2013 IEEE Antennas and Propagation Society International Symposium (APSURSI), Orlando, FL, USA, 2013, pp. 1614-1615

[5] Metastatic,

Last accessed: 01/31/2023

[6] Jouhari, M., Amhoud, E. M., Saeed, N., and Alouini, M.-S., “A Survey on Scalable LoRaWAN for Massive loT: Recent Advances, Potentials, and Challenges”, 2022. doi: 10.48550/arXiv.2202.11082.

[7] Samtech,

Last accessed: 01/31/2023

[8] Q. Liu, X. J. Bai, X. Gan, and S. Yang, “LoRa RTT ranging characterization and indoor positioning system,” Wireless Communications and Mobile Computing, 2021

[9] F. Rander Andersen, K. Dilip Ballal, M. Nordal Petersen and S. Ruepp, "Ranging Capabilities of LoRa 2.4 GHz," 2020 IEEE 6th World Forum on Internet of Things (WF- loT), New Orleans, LA, USA, 2020, pp. 1-5

[10] Samtech, “Application Note: An Introduction to Ranging with the SX1280 Transceiver”, Mar. 2017

[11] P. Muller, H. Stoll, L. Sarperi and C. Schupbach, "Outdoor Ranging and Positioning based on LoRa Modulation," 2021 International Conference on Localization and GNSS (ICL-GNSS), Tampere, Finland, 2021, pp. 1-6, doi: 10.1109/ICL- GNSS5 1451.2021.9452277.

[12] Semtech Corporation, "Application Note: An Introduction to Ranging with the

SX1280 Transceiver," Accessed: Feb. 02, 2021. [Online], Available: https://www.semtech.com/products/wireless-rf/24-ghz-transceivers/sxl280

[13] S. Lanzisera, D. T. Lin, and K. S. J. Pister, "RF Time of Flight Ranging for Wireless Sensor Network Localization," 2006 International Workshop on Intelligent Solutions in Embedded Systems, Vienna, Austria, 2006, pp. 1-12, doi: 10.1109/WISES.2006.329127.

Claims

1. A method for exchanging data from under a collapsed building, characterized by:

- providing multiple communication nodes (N) placed inside the building and at least one communication node (RN) positioned outside the rubble before the building collapses,

- for the autonomous establishment of said communication nodes, i. the communication nodes acting as an initiator or responder periodically during (N) T_period and being able to reach the communication channel actively during initiator and the said communication nodes acting as an initiator with the probability p E [0,1], ii. Selecting the most suitable slots by determining how many their own and neighboring nodes use the same slot as their neighbors when using a slot for the communication nodes (N) provided as integers,

- locating the position of a communication node (N) in said rubble using at least one communication node (N) positioned outside the wreckage with a known position and capable of communicating with a communication node (N) trapped in the rubble and progressively locating the position of another communication node (N) under rubble using the communication node (N) with a known position,

- sending data outside the rubble using communication nodes (N).

2. A method according to claim 1, characterized in that a translation time is applied to align said communication nodes (N) as initiators.

3. A method according to claim 2, characterized in that the runtime of a node with a translation time applied as the initiator is determined according to the following formula:

and wherein,

is the local time information of the node,

is the translation time and is the slot where the node takes the initiator role, and n is the

parameter indicating which node the node is.

4. A method according to claim 1, characterized in that, in order to reduce slot conflicts,

- communication nodes (N) broadcast the identity and slot information of its detected neighbors by the with the probability of p when they pass the initiator state,

- when a communication node (N) receives announcement information from the other communication node (N), if the slot used by itself is not listed as a neighbor among the slots used by the communication node (N) or its neighbors sending the packet specified in the announcement packet, it randomly selects a slot if there is a non-conflicting slot or use the same slot if there is no conflict.

5. A method according to claim 4, characterized in that it selects the translation parameter as according to the announced slot of the incoming

packet and the time of its slot, and wherein refers to the slot of the communication

node receiving the announcement information and, is the slot of the announcing

communication node (N).

6. A method according to claim 1, characterized in that the round-trip time-of-flight method is used to find the positions of the communication nodes (N) under the rubble.

7. A method according to claim 6, characterized in that: the round-trip time-of-flight method and a communication node (N) send a ranging request packet of length A_req in discrete time, specifying the address of a particular communication node (N) as the initiator, the initiator communication node (N) starts to count the ticks of its clock at a f_A frequency at this stage, the communication node (N) receiving the ranging request packet sends the ranging response packet after a certain time and reaches the initiator communication node (N) after T_p seconds; and

T_p value is calculated as follows:

and the distance determination is performed using the value T_p.

8. A method according to claim 6, characterized in that it comprises: a communication node (N) sends the start-ranging packet between these two communication nodes (N) by specifying the address of another communication node (N), the communication node (N), after receiving the start-ranging packet, changes its mode to ranging responder and waits for the ranging request packet, the communication node (N) sending the start-ranging packet broadcasts the ranging request packet and the communication node (N) receiving this packet broadcasts the ranging response packet and the other communication node (N) receives the packet and records the data.

9. A method according to claim 1, characterized in that communication outside the rubble is carried out with a minimum number of communication nodes (N).

10. A method according to claim 1, characterized by, in order to realize communication outside the rubble, the h_n value indicating the number of steps the communication node (N) has taken to determine the position during the progressive position discovery is defined and the communication node (N) to transmit the message transmits the message to the communication node (N) with the lowest h_n value it can communicate with, and repeats this method until the message reaches outside the rubble.

11. A method according to claim 1, characterized in that, in order to realize communication outside the rubble, the h_n value indicating at which step the communication node (N) reaches the reference node (RN) is defined and the communication node (N) to transmit the message transmits the message to the communication node (N) with the lowest h_n value it can communicate with, and repeats this method until the message reaches outside the rubble.

12. A method according to any one of the preceding claims, characterized in that said communication nodes (N) are long distance nodes.

13. A method according to claim 12, characterized in that the communication nodes (N) operate in the Long-Range Wide Area Network standard.

14. A system comprising communication nodes having multiple and at least one processing unit capable of executing a method according to any one of claims 1-13 for exchanging data from a collapsed building.