2018 IEEE 16th Int. Conf. on Dependable, Autonomic & Secure Comp., 16th Int. Conf.

on Pervasive Intelligence &

Comp., 4th Int. Conf. on Big Data Intelligence & Comp., and 3rd Cyber Sci. & Tech. Cong.

Wi-Fi Halow Signal Coverage Estimation in

Collapsed Structures
Muhammad Faizan Khan† , Guojun Wang∗† , Md Zakirul Alam Bhuiyan‡ , and Tao Peng†
† School of Computer Science and Technology, Guangzhou University, Guangzhou, China, 510006
‡ Department of Computer and Information Sciences, Fordham University, New York, NY, 10458

∗
Correspondence to: [email protected]

Abstract—With the growing research for IoTs particularly Wi-Fi Halow is primarily for IoTs and has good penetration
wireless signals as sensors, Wi-Fi has witnessed an increasing for complex indoor environments. This is because of its
trend for numerous applications.In this study, we investigate operation at lower frequency i.e., 900MHz. So, having this
the feasibility of Wi-Fi Halow signals for collapsed structured
environments. The objectives of this paper are first to consider a understanding, we envisage that these signals can be effective
complex debris scenario and then wireless signal coverage under for collapsed structures where we can rescue lives. This can
this debris has to be computed. A modified path loss model known turn into complete ubiquitous post-disaster rescue solution
as PL-Collapsed along with complex collapsed structure layout with device free communication because of Wi-Fi Halow
has been used for proper reception of echo which provides the replacing the traditional IoTs where traditionally, there is need
probable coverage for aforementioned scenario. Objectives have
been realized with proper simulations. Then, we compare the of cyber-physical systems or robots.
results for various signal intensities, antenna gains and debris In this paper, we employ a compendious model PL- Col-
layers. Comparison of results show that Wi-Fi Halow at low lapsed for path losses in collapsed structures which ensures
power with higher antenna gains has better coverage. It is effective computation of coverage of weak Wi-Fi signals.
therefore paving way for post-disaster rescue through IoTs in
comparison with traditional radar techniques.
In our approach, we also consider debris model which is
Index Terms—Wi-Fi Halow, coverage, IoT, collapsed structure, constituent of brick, concrete, glass and lumber as materials.
path loss The objective of this study is to estimate the coverage of
Wi-Fi signal through our debris model while employing PL-
I. I NTRODUCTION Collapsed. The placement of transmitter and receiver i.e.,
Wi-Fi is considered to be outside of debris to model it
With the advent of internet of things (IoTs), our lives are like a real environment. The transmitter continuously sends
becoming more dependent on ubiquitous systems [1] [2]. Sen- signals like Doppler radar which pass through debris and
sors and wearables are essential part of IoTs but now research face attenuation because of complexity of layout, materials
has much focused on device free communication [3] [4]. This and causes multipath. This multipath fading, attenuations,
essentially include wireless devices at a low power to be acting reflections and scattering weakens the already weak Wi-Fi
as sensors particularly Wi-Fi signals that have additional roles signals. These signals reflect back to receiver after facing so
than sole Internet connectivity medium. [5] [6]. These signals much attenuations from debris model and provide coverage
are very much effective for indoor environments where mostly information. We have modelled collapsed structure having
IoTs are deployed [7]. specification of brick 10.5´´, concrete 8´´, glass (6mm), glass
Application of Wi-Fi for indoor environments has received (13mm) and lumber (76mm). These specifications have been
much attention because of complexity and nature of civil considered based on construction style of developing coun-
structures where wireless signals pass through direct or multi- tries. We have also considered minimum detectable signal
path. It is also because of global positioning system which strength which guarantees the quality of echo as it contains
is not suitable for indoor scenarios [7] [8]. Wireless signals all vital information for rescue [5].
convey information that typifies the environment when these The main contributions of this paper are:
pass through indoor physical spaces [5]. This becomes more
troublesome in case of collapsed structure where multiple ob- • We envisage the PL-Collapsed for attenuation and path
jects constituent of various materials cause more signal distor- losses for brick and concrete collapsed structures.
tion eventually completely fading the wireless signals [9] [10]. • We make an investigation of Wi-Fi Halow signal behavior
And penetration of wireless signals under collapsed structure for collapsed structure through our debris model.
is more important because it can lead to post disaster rescue • Finally, we validate our model for low power Wi-Fi
as existing solution are expensive and are not available in Halow. This work highly emphasizes on future IoTs
developing countries where most of the collapses occur. which would be based on sensor less sensing.
The main objective of this paper is to focus on feasibility of We organize this paper as follows: Section II provides more
Wi-Fi Halow signals and its coverage for collapsed structures. depth on background and related work whereas Section III de-

978-1-5386-7518-2/18/$31.00 ©2018 IEEE 626

DOI 10.1109/DASC/PiCom/DataCom/CyberSciTec.2018.00113
scribes our problem formulation. We discuss our modified path of radar used in rescue efforts are Doppler radars [24] [25]
loss model for coverage in Section IV. Simulation results have [26], Ultra-wideband(UWB) radars [27], IR-UWB radars [28]
been provided in Section V. Section VI completely analyzes and Continuous Wave Stepped Frequency (CW-SF) Ground
the work and compare it prior works. Finally, Section VII penetrating radar [29]. These approaches work on different
concludes the paper. frequencies ranging from lower to higher such as 270MHz,
400MHz [28] to 1.15GHz [26]. These technologies are also
II. BACKGROUND AND R ELATED W ORK expensive and their availability in developing countries is
In this section, we provide deep insight to significance of still a question. These technologies are also not feasible for
this work by articulating the background and related work. complex structures hence limiting their applications.
Firstly, factors pertaining to wireless signal under collapsed In a nutshell, we observe that most of the current rescue
structure are presented. These include path loss modelling solutions using wireless signals are mainly based on radars
and attenuations under debris. Afterwards, applications of and cannot be considered as ubiquitous. These solutions are
wireless signals using various technologies for rescue have quite expensive and also not available in third world countries
been discussed. Finally, we provide the problem statement where most of the collapses occur. As to date, there is no
based on this background and related work. ubiquitous solution particularly Wi-Fi Halow based one for
Wireless signals face reflections, attenuations and scattering coverage estimation in collapsed structure which can maximize
on being exposed to objects in their path of communication. rescue efforts. In the light of related work, we are certain that
This weakens the wireless signals and range is highly affected. investigation of Wi-Fi Halow for collapsed structures if proved
Considering the collapsed structure environments, it becomes to be successful can make a turnaround in ubiquitous solutions.
more cumbersome as there are layers of objects laying over
each other hence forming a debris. So, these attenuations III. P ROBLEM F ORMULATION
caused by scattering and losses make it difficult for wireless In this section, we present problem formulation. We firstly
signal to penetrate through debris [9]. DiCarlofelice and E. Di- investigate the wireless signal behavior in collapsed structures
Giampaolo [10] [11] conducted localization of radio emitters with an echo model followed by debris model which simplifies
into collapsed buildings after an earthquake. They conducted our problem.
a measurement campaign in a typical European historical
city, LAquila, which was stroke by a severe earthquake [5]. A. Wi-Fi Halow Signal and Echo Model
Identification of best radio band for signal penetration under Indoor environments have relatively complex structures as
debris is subject of great importance. A study was conducted these are constituent of multiple objects, walls, halls etc., so
by [9] for 50 MHz, 150MHz, 225MHz, 450MHz, 900MHz, wireless signal faces many attenuations and becomes weak.
and 1.8GHz. Similarly, there has been research conducted When it comes to collapsed structure, the case becomes more
on structural health monitoring as in [12] [13] [14], which worse as even single wall may turn up to multiple objects. It
also provide intuition about wireless signal applications for is evident that by considering whole collapse structure where
complex environments. debris has hundreds of pieces, wireless signal penetration
The fundamental challenge to deal with attenuations and becomes a nightmare. And it becomes even worse for weak
losses encountered to wireless signals under collapsed struc- Wi-Fi signal to penetrate into it.
ture is effective path loss modeling. Unfortunately, there is In order to simplify the problem, let us consider a small
very few significant work done for path loss computation in indoor collapsed structure consisting of brick, concrete, glass
context of this scenario as [15]. Although, there is enough and lumber only as shown in Fig 1a. This shows that aforemen-
literature available for path loss modelling in other outdoor tioned structure has turned into multi object scenario where
and indoor environments which provide us insight towards each object having different thickness level which defines the
their application in collapsed cases. A systematic survey of Wi-Fi signal penetration.
path loss prediction methods, covering more than 60 years In order to address the coverage issue, we employ echo
of incessant research has been presented in [7] [16]. Rap- model which works like Doppler radar. Wi-Fi transmitter
paport [8], firstly proposed path loss modelling for complex radiates its signals simultaneously that can take many paths but
environments. Afterwards, many researchers put emphasis in for simplicity of problem, we have only considered directional
this domain with studies like [17] [18]. There are also some lobe of antenna. When this signal passes through collapsed
studies related to Wi-Fi bands such as [5] [19] and [20]. structure as model shown in Fig 1b, it will face multiple
Finally, we discuss existing techniques that employ wire- attenuations and fading. These fading will be of every type
less signals for possible rescue under collapsed environment. like multipath fading to shadowing, fast fading to slow fading
Although, there are several tools and technologies available which in turns weaken the signal. This weak signal will return
for detection of human activity under debris such as sound back after facing so much reflections.We properly consider
sensor [21], micro-cameras [22],wireless sensors [23] and MDS for wireless signals to ensure quality reception of echo
radars [24] but the most favored method is radar based [5].
sensing. It is because of the fact that it has better ability Normal Wi-Fi signals operate at higher frequencies i-e,.
to penetrate deep through dielectric barriers. The most types 2.4GHz and 5GHz and are more prone to attenuations. On

627
 (a) Simple Collapsed Structure (b) Signal and Echo Model (c) Minimum Layer Case (d) Maximum Collapse Case

Fig. 1: Illustration of collapsed structure with echo and layered models

the other hand, Wi-Fi Halow is latest technology operating transceivers to compute coverage. Path losses incurred to Wi-
at sub 1GHz and is specifically designed for obstructions Fi signal in collapse structures define the validity of model.
and attenuations. So, in our signal model, we consider Wi-Fi The placement of Wi-Fi Halow accentuates the signal pen-
Halow for collapsed structures as it has good penetration for etration in collapsed structures. We assume that directional
obstructions and can pave way for ubiquitous solution in post- lobe of antenna is only the main lobe in order to avoid mul-
disaster rescue as there is enough research done to estimate tiple aliased signal reflected from undesired sources. We also
the breathing behavior on normal Wi-Fi band in normal indoor assume that there is no other transmitter and no other sources
environments [30] [31]. of noises. Wi-Fi signal penetration in collapsed environments
can be realized by considering link budget equation given as;
B. Debris Model
Selection of proper debris is pivotal to estimate the cover- Prx = Ptx + Gtx + Grx − P L (1)
age. Collapsed structures may have multi-material debris but where G and P is antenna gains and power for transceivers.
here we only consider brick, concrete, glass and lumber based The term P L defines path losses incurred to signal because of
debris. We assume that after collapse, structure turn up into debris. It becomes more apparent when signal tries to deeply
pieces laying over each other. We also assume that collapse is penetrate into debris with the increasing distance.
caused by an earthquake. We define layers based on this fact Wireless signal propagation for our debris model will con-
of pieces laying over each other and make a debris model like sist of two parts. Firstly, it will travel in free-space and then
in Fig 1c. Debris has vertical and horizontal layers but more it will penetrate through obstructed paths. This can be given
important are vertical layers as these define the depth of debris. as below;
The signal penetration of horizontal and vertical layers will be
almost same but in order to address the most important depth P L = P Lf ree−space + P Ldebris (2)
issue, we only consider vertical layers. All debris materials Firstly, we analyze the path loss in free space. We assume
as considered above have their own attenuation behavior for that Wi-Fi signal will travel only 1 meter before entering to
various frequencies. We have considered both normal depth debris. We also make assumption that there is no other obstacle
and deep depth cases with Fig 1d showing the common depth between Wi-Fi signal transmitter to debris. This signal will
scenario whereas Fig 1c depicting complex debris layers. It is penetrate through our debris model after traveling through
evident that each adding piece will cause more obstructions, free-space where it will face much attenuations primarily be-
so deeper the debris, higher the attentuation which results in cause of debris layers formed after collapse laying horizontally
lower penetration. or vertically. Path loss in debris can be derived using distance
So, in brief, our problem is to address coverage (which can dependent path formula. It states that path loss increases
further assist in rescue) in collapsed structure based on MDS exponentially with the distance given in log form as below;
threshold [5]. This coverage various with respect to transmitter
d
intensities, antenna gains, minimum detectable signal thresh- P Ldebris = P L(d0 )[dB] + 10 ∗ α ∗ log10 ( ) (3)
old and debris model. We estimate the coverage with these d0
aforementioned aspects and predict the Wi-Fi Halow signal where P Ldebris is path loss in debris and P L(d0 ) is
behavior for collapsed structures. free space path loss for d0 . There is no P L(d0 ) in debris,
so, free-space path loss is only computed for outside space
IV. A PPROACH : PL-C OLLAPSED of debris. α is path loss exponent which also varies with
In this section, we present PL-Collapsed. This model environments ranging from 2 for free-spaces to 6 in case of
considers echo and debris model along with geometry of obstacles consisting metallic objects as well. We assume it to

628
be 5 as there is no high attenuation metallic source under TABLE I: Simulation Parameters
consideration. Now, in order to incorporate attenuations of
multiple layers of debris, ( 2) can be rewritten as;
Parameter Values
d
P L = P Lf ree−space + 10 ∗ α ∗ log10 ( ) + Xg (4) Operation Frequency (902, 915,917)MHz
d0
Transmission Power (30, 24, 23, 13, 10) dBm
Xg is new term introduced here which includes all possible Antenna Gains (6, 12, 13, 23, 26) dBi
losses. Wi-Fi signal faces much attenuation because of reflec- Debris Type Brick, Concrete, Glass, Lumber
Attenuation of Brick 10.5´´ 7dB at 900 MHz
tions, scattering, multiple paths, diffractions and fading. The Attenuation of Concrete 8´´ 23dB at 900 MHz
most evident form of fading in our case are both slow-fading Attenuation of Glass (6mm) 0.8dB at 900 MHz
and fast-fading but for simplicity of approach, we have only Attenuation of Glass (13mm) 2dB at 900 MHz
Attenuation of Lumber (76mm) 2.8dB at 900 MHz
exploited attenuation factor without going into modification of Minimum Debris Layers 15 (4b+2c+6g+3G+2l)
those models. ( 2) with addition of possible attenuation will Maximum Debris Layers 34 (10b+4c+10g+6G+4l)
be as follows; Modulation Type 256QAM, 16QAM
Channel BW at 256QAM 1MHz
d Channel BW at 16QAM 16MHz, 1MHz
P L = P Lf ree−space + 10 ∗ α ∗ log10 ( ) MDS Threshold at 256QAM -70dBm
d0 MDS Threshold at 16QAM -71dBm, -83dBm
+ l ∗ AF (Bricklayer) + m ∗ AF (ConcreteLayer) Path Loss Exponent 5
+ n ∗ AF (Glass(6mm)Layer)
+ o ∗ AF (Glass(13mm)Layer)
+ p ∗ AF (Lumber(76mm)Layer) Japan(915MHz) and Korea(917MHz). These countries have
(5) their own spectrum limits and power considerations for Wi-Fi
where l, m, n, o, and p denotes number of layers for our Halow as in [32]. Similarly variance in transmission power
considered materials. AF is attenuation factor caused by and antenna gains is regulated with FCC para 17.245 rules for
single layer of brick, concrete, glass and lumber respectively. PTMP (Point to multi-point). Antenna gain varies from 6dBi to
Thickness of materials will also define the complexity of layers 26dBi in relation with transmitted power. Wi-Fi Halow follows
and attenuation values vary accordingly. same behavior for power increase in relation with decrease in
Finally, we consider the echo signal which has desired antenna gain to that of 2.4GHz and 5GHz spectrum as given
information for rescue. It means the strength of received signal in regulations2 .
which conveys proper information about breathing signals As mentioned earlier that collapsed structure may constitute
under debris. Wi-Fi signals have feature known as channel various materials. We have considered four types of debris
state information(CSI) which contains the sensing information. constituent’s based on real construction scenarios in devel-
This CSI can further help in post-disaster rescue. In order to oping countries. These are brick´´, concrete8´´, glass(6mm),
achieve required echo, the received power should be above glass(13mm) and lumber(76mm). The depth of these materials
or equal to pre-defined threshold termed as MDS. It can be define the complexity of collapsed structures. The attenua-
defined for a given radio as a function of acceptable error rate tions caused by single layers of under consideration materials
M DS(Pe ) (i.e., probability of acceptable error rate; Pe ) can have been provided by Digi 3 . We also select debris layers
be given as below; which defines the nature of collapsed structure. Most of
the buildings in developing countries are not sky scrappers;
Ptx + Gtx + Grx − P L ≥M DS(Pe ) (6) so, we have assumed debris layers between 15 to 34 for
V. S IMULATIONS simulations in order to have insight about Wi-Fi Halow Signal
in collapsed environments. The minimum detectable signal
This section discusses the simulation method and provide
threshold which is much important for better echo reception;
results. Simulation parameters vary with the environments
specifically for our case as Quadrature amplitude modula-
primarily because of regulatory bodies like Federal Commu-
tion(QAM); is assumed to be -70dBm for 256QAM case with
nications Commission (FCC)1 .
1MHz channel bandwidth. Similarly, for lower modulation
A. Simulation Method scheme i.e., 16QAM, -71dBm and -83dBm values have been
Right selection of simulation parameters is essential to have considered for MDS threshold with 16MHz and 1MHz channel
better results. We have performed simulations on MATLAB bandwidth respectively. Higher QAM ensures better data rate
17a on 64bit OS with 16GB RAM and Intel(R) Xeon(R) 3.3 but noise and interference are also evident. That is why, we
GHz CPU as per simulations parameters shown in Table I. have considered both low and high QAM case to have better
The operating frequency for Wi-Fi Halow is provided understanding of Wi-Fi Halow. Finally, we select path loss
in Table I, which is regulated by IEEE Task Group [32]. exponent as 5 in order to cope with cluttered environment.
We have considered the frequency case for US(902MHz),
2 https://www.air802.com/fcc-rules-and-regulations.html
1 https://www.fcc.gov/tags/radio-rules
3 http://ftp1.digi.com/support/images/XST-AN005a-IndoorPathLoss.pdf

629
 TABLE II: Coverage Range with Various Materials

Transmit Max Brick Concrete Glass Lumber Range in Collapsed Structure (m)
Power Antenna Attenuation Attenuation Attenuation Attenuation
(dBm) (dBi) 10.5´´ 8´´ 76mm
(dB) (dB) 6mm 13mm (dB) 256QAM (1MHz) 16QAM (16MHz)
(dB) (dB)
Min L Max L Min L Max L

4b+2c+6g 10b+4c+10g 4b+2c+6g 10b+4c+10g

+3G+2l +6G+4l +3G+2l +6G+4l

30 6 0.5502 0.0048 0.5761 0.0051

24 12 1.2605 0.0111 1.3199 0.0116
23 13 1.4472 0.0127 1.1514 0.0133
7 23 0.8 2 2.8
256QAM (1MHz) 16QAM (1MHz)

13 23 5.7234 0.0503 10.4149 0.0915

10 26 8.6552 0.0761 15.7498 0.1384

(a) 256QAM, 1MHz Channel (b) 16QAM, 16MHz Channel

Fig. 2: 30dBm, 6dBi Scenario

(a) 256QAM, 1MHz Channel (b) 16QAM, 16MHz Channel

Fig. 3: 24dBm, 12dBi Scenario

B. Simulation Results across different debris layers with varied modulation types
We performed simulation on afore-discussed parameters. and channel consideration. The distance power relationship
Results as displayed in Table II have shown an increasing for these intensities has been shown in Fig 2 to Fig 6. It
trend of coverage with lowering the transmitter intensity. can be observed that higher transmission power provides low
Table II also depict Wi-Fi Halow signal frequency behavior coverage, e.g 30dBm power with 256QAM covers 0.5502m
for different debris layers across varied signal intensities and and 0.0048m respectively for min and max layer scenario in
antenna gains. Overall, we performed 20 simulations with debris. On the contrary, 10dBm power with 256QAM covers
varied parameters. Finally, we compare the results for best 8.6552m and 0.0761m respectively for same debris type. This
power level which should be low enough to fit it for IoT based make us believe that Wi-Fi Halow operating at lower power
solutions. provide better coverage. This comparison has been illustrated
1) Comparison w.r.t Signal Intensity: Let us compare sim- in Fig 7, where an increasing signal coverage can be seen with
ulation results for various Wi-Fi Halow signal intensities lower power, higher antenna gain and lower QAM type.

630
 (a) 256QAM, 1MHz Channel (b) 16QAM, 16MHz Channel
Fig. 4: 23dBm, 13dBi Scenario

(a) 256QAM, 1MHz Channel (b) 16QAM, 16MHz Channel

Fig. 5: 13dBm, 23dBi Scenario

(a) 256QAM, 1MHz Channel (b) 16QAM, 16MHz Channel

Fig. 6: 10dBm, 26dBi Scenario

(a) Comparison w.r.t modulation scheme (b) Stacked graph comparison

Fig. 7: Comparison of Trans Power, Antenna Gain and Coverage

631
 2) Comparison w.r.t Signal Modulation Type: In this sub- and poor management. Considering the ubiquitous solutions
section, we compare Wi-Fi Halow signal coverage on the converging in daily life, we are certain that there is also need
basis of channel bandwidth and modulation coding scheme to investigate IoT based solutions for rescue. Wi-Fi signals are
being considered. Both cases of debris layers are tested for available in almost every country and recently presented Wi-Fi
256QAM with 1MHZ and 16QAM with (16MHz, 1MHz) Halow operating at lower frequency can realize our goal.
channel bandwidths respectively. Wi-Fi Halow frequencies for We mostly prefer power efficient IoT devices. In this work,
US, Korea and Japan operate in 1MHz for QAM256 but differ we have simulated Wi-Fi Halow across different power level
with 16QAM, where these still operate in 1MHz channel and figured out that low transmission power provides us better
bandwidth. Let us consider 24dBm power as an example results. Secondly, it is operating at sub 1GHz frequency in
for comparison of coverage for modulation coding schemes. comparison with traditional Wi-Fi signals which operate at
Simulation results for this case are shown in Fig 3a and 3b 5GHz and 2.4GHz respectively. We have considered modified
respectively. It can be observed from these results that 16QAM path loss model similar to [7] [8] which take complex structure
perform better than higher QAM value i.e.,256QAM. We il- into consideration. If we compare this work with existing path
lustrate overall comparison of coverage on base of modulation loss models, it can be concluded that most of the works like [9]
type in Fig 7. [17] [20] [33] [34] provide path loss computation for various
3) Comparison w.r.t Debris Type: Finally, we compare the environments but none of these deal with collapsed structures.
Wi-Fi Halow signal behavior for our proposed debris cases. Finally, we discuss the possible application of low power
We observe that minimum layer scenario perform better than of Wi-Fi Halow for collapsed environments. In order to have
maximum layer. This can be seen in all simulation results efficient rescue, we need to fulfil two conditions. These are
across various signal intensities and modulation schemes from breathing or heart beat detection and localization of alive
Fig 2 to Fig 6. In order to show better performance, we one under debris. Interestingly, there are researches on the
compare 10dBm power case as shown in Fig 6. The coverage application of Wi-Fi signals for breathing detection [35] and
for QAM256 under min layer scenario is 8.6552m, whereas localization [36] [37] [38] at 2.4GHz. So, in order to have
for max layer; it is 0.0761m. It is quite evident from both ubiquitous solution, there is need to map both applications
values that signal performs better in lower layered debris. under debris scenario where 2.4GHz performs poorly but Wi-
Similarly, for same power level but with QAM16, the coverage Fi Halow provide better results. So, we can rightly say that
distance for min layer has risen to 15.7498m in contrast with research on breathing and localization using Wi-Fi Halow
max layer case, where it is 0.1384m. So, it can be concluded signals in collapsed environments may revolutionize the rescue
that increasing the debris layers will cause more attenuations operations in the future.
for which we need to increase signal strength through various
means in order to have deep coverage. VII. C ONCLUSION
In this paper, the application of Wi-Fi Halow for col-
VI. P ERFORMANCE A NALYSIS lapsed structure was under consideration. Wireless signals are
In this section, a deep analysis of our work is provided severely affected in collapsed environments. It is because of
with proper background, contributions and insight for future. complex debris which causes higher attenuation. This work
Firstly, there is need to identify the complexity of problem. dealt with the behavior of Wi-Fi Halow signal at different
Afterwards, contributions and comparison with existing works intensities, geometry of transceivers and complexity of debris
in relevant domain is to be compared for justification of in order to have better coverage which may return echo
this work. Finally, a discussion is made on importance and with desired information. Future research can be carried on
application of Wi-Fi Halow for possible rescue as a future increasing the coverage distance under collapsed environment
work. which can then further trigger post-disaster rescue with IoTs.
In this work, we estimated Wi-Fi Halow signal coverage
ACKNOWLEDGMENT
in collapsed structures. These environments are very complex
in nature as buildings after collapse form multiple layers of National Natural Science Foundation of China under Grants
debris depending on the type of collapse, causes and buildings 61632009 and 61472451,Guangdong Provincial Natural Sci-
structural constituent. This debris makes multi-layered case ence Foundation under Grant 2017A030308006 High-Level
where wireless signals behave extremely poor. This is one Talents Program of Higher Education in Guangdong Province
of the main reason that still manual rescue is going on under Grant 2016ZJ01 [5].
around the world in case of collapses. There is very little
R EFERENCES
work done on the application of wireless signals for collapsed
environments. Most of these studies like [11] [24] [29], make [1] J. Lin, W. Yu, N. Zhang, X. Yang, H. Zhang, and W. Zhao, “A survey
on internet of things: Architecture, enabling technologies, security and
use of radar based approach but could not provide desired privacy, and applications,” IEEE Internet of Things Journal, vol. 4, no. 5,
results except providing insight about application of wireless pp. 1125–1142, oct 2017.
signals in collapsed structures. These techniques are also costly [2] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, and
M. Ayyash, “Internet of things: A survey on enabling technologies,
and can not be deployed in developing countries where most protocols, and applications,” IEEE Communications Surveys & Tutorials,
of the collapses occur because of low engineering standards vol. 17, no. 4, pp. 2347–2376, 2015.

632
 [3] M. Zhao, F. Adib, and D. Katabi, “Emotion recognition using on PErvasive Technologies Related to Assistive Environments, ser.
wireless signals,” in Proceedings of the 22Nd Annual International PETRA ’17. New York, NY, USA: ACM, 2017, pp. 311–318.
Conference on Mobile Computing and Networking, ser. MobiCom ’16. [Online]. Available: http://doi.acm.org/10.1145/3056540.3076201
New York, NY, USA: ACM, 2016, pp. 95–108. [Online]. Available: [23] J. Wang, Z. Cheng, L. Jing, and T. Yoshida, “Design of a 3d localization
http://doi.acm.org/10.1145/2973750.2973762 method for searching survivors after an earthquake based on wsn,” in
[4] P. J. Soh, G. A. Vandenbosch, M. Mercuri, and D. M.-P. Schreurs, 2011 3rd International Conference on Awareness Science and Technol-
“Wearable wireless health monitoring: Current developments, chal- ogy (iCAST), Sept 2011, pp. 221–226.
lenges, and future trends,” IEEE Microwave Magazine, vol. 16, no. 4, [24] L. Crocco and V. Ferrara, “A review on ground penetrating radar
pp. 55–70, may 2015. technology for the detection of buried or trapped victims,” in 2014
[5] M. F. Khan, G. Wang, M. Z. A. Bhuiyan, and X. Li, “Wi-fi signal International Conference on Collaboration Technologies and Systems
coverage distance estimation in collapsed structures,” in 2017 IEEE (CTS), May 2014, pp. 535–540.
International Symposium on Parallel and Distributed Processing with [25] R. M. Narayanan, “Earthquake survivor detection using life signals from
Applications and 2017 IEEE International Conference on Ubiquitous radar micro-doppler,” in Proceedings of the 1st International Conference
Computing and Communications (ISPA/IUCC). IEEE, dec 2017. on Wireless Technologies for Humanitarian Relief, ser. ACWR ’11.
[6] Z. Zhou, C. Wu, Z. Yang, and Y. Liu, “Sensorless sensing with wifi,” New York, NY, USA: ACM, 2011, pp. 259–264. [Online]. Available:
Tsinghua Science and Technology, vol. 20, no. 1, pp. 1–6, Feb 2015. http://doi.acm.org/10.1145/2185216.2185288
[7] C. Phillips, D. Sicker, and D. Grunwald, “A survey of wireless path [26] F. JalaliBidgoli, S. Moghadami, and S. Ardalan, “A compact portable
loss prediction and coverage mapping methods,” IEEE Communications microwave life-detection device for finding survivors,” IEEE Embedded
Surveys Tutorials, vol. 15, no. 1, pp. 255–270, First 2013. Systems Letters, vol. 8, no. 1, pp. 10–13, March 2016.
[8] S. Y. Seidel and T. S. Rappaport, “914 mhz path loss prediction models [27] J. Li, L. Liu, Z. Zeng, and F. Liu, “Advanced signal processing for
for indoor wireless communications in multifloored buildings,” IEEE vital sign extraction with applications in uwb radar detection of trapped
Transactions on Antennas and Propagation, vol. 40, no. 2, pp. 207– victims in complex environments,” IEEE Journal of Selected Topics in
217, Feb 1992. Applied Earth Observations and Remote Sensing, vol. 7, no. 3, pp. 783–
[9] C. L. Holloway, G. Koepke, D. Camell, W. F. Young, and K. A. Remley, 791, March 2014.
“Propagation measurements before, during, and after the collapse of [28] Z. Li, H. Lv, Y. Zhang, G. Lu, S. Li, X. Jing, and J. Wang, “Detection
three large public buildings,” IEEE Antennas and Propagation Maga- of trapped survivors using 270/400 mhz dual-frequency ir-uwb radar
zine, vol. 56, no. 3, pp. 16–36, June 2014. based on time division multiplexing,” in 2014 IEEE Topical Conference
[10] A. D. Carlofelice, E. D. Giampaolo, M. Elaiopoulos, M. Feliziani, on Biomedical Wireless Technologies, Networks, and Sensing Systems
M. Roselli, and P. Tognolatti, “Localization of radio emitters into (BioWireleSS), Jan 2014, pp. 31–33.
collapsed buildings after earthquake: Measurements of path loss and [29] G. Grazzini, M. Pieraccini, F. Parrini, A. Spinetti, G. Macaluso, D. Dei,
direction of arrival,” in International Symposium on Electromagnetic and C. Atzeni, “An ultra-wideband high-dynamic range gpr for detecting
Compatibility - EMC EUROPE, Sept 2012, pp. 1–6. buried people after collapse of buildings,” in Proceedings of the XIII
[11] A. DiCarlofelice, E. DiGiampaolo, M. Feliziani, and P. Tognolatti, Internarional Conference on Ground Penetrating Radar, June 2010, pp.
“Experimental characterization of electromagnetic propagation under 1–6.
rubble of a historic town after disaster,” IEEE Transactions on Vehicular [30] X. Liu, J. Cao, S. Tang, J. Wen, and P. Guo, “Contactless respiration
Technology, vol. 64, no. 6, pp. 2288–2296, June 2015. monitoring via off-the-shelf wifi devices,” IEEE Transactions on Mobile
[12] M. Z. A. Bhuiyan, G. Wang, J. Wu, J. Cao, X. Liu, and T. Wang, “De- Computing, vol. 15, no. 10, pp. 2466–2479, Oct 2016.
pendable structural health monitoring using wireless sensor networks,” [31] J. Liu, Y. Wang, Y. Chen, J. Yang, X. Chen, and J. Cheng,
IEEE Transactions on Dependable and Secure Computing, vol. 14, no. 4, “Tracking vital signs during sleep leveraging off-the-shelf wifi,”
pp. 363–376, July 2017. in Proceedings of the 16th ACM International Symposium on Mobile
[13] M. Z. A. Bhuiyan, J. Wu, G. Wang, Z. Chen, J. Chen, and T. Wang, Ad Hoc Networking and Computing, ser. MobiHoc ’15. New York,
“Quality-guaranteed event-sensitive data collection and monitoring in NY, USA: ACM, 2015, pp. 267–276. [Online]. Available: http:
vibration sensor networks,” IEEE Transactions on Industrial Informatics, //doi.acm.org/10.1145/2746285.2746303
vol. 13, no. 2, pp. 572–583, April 2017. [32] “Ieee draft standard for information technology - telecommunications
[14] M. Z. A. Bhuiyan, J. Wu, G. Wang, T. Wang, and M. M. Hassan, and information exchange between systems - local and metropolitan area
“e-sampling: Event-sensitive autonomous adaptive sensing and low- networks - specific requirements - part 11: Wireless lan medium access
cost monitoring in networked sensing systems,” ACM Transactions on control (mac) and physical layer (phy) specifications: Amendment- sub
Autonomous and Adaptive Systems, vol. 12, no. 1, pp. 1–29, mar 2017. 1 ghz license-exempt operation,” IEEE P802.11ah/D2.0, June 2014, pp.
[15] L. Chen, M. Loschonsky, and L. M. Reindl, “Large-scale fading model 1–582, April 2014.
for mobile communications in disaster and salvage scenarios,” in 2010 [33] C. R. Anderson and T. S. Rappaport, “In-building wideband partition
International Conference on Wireless Communications Signal Process- loss measurements at 2.5 and 60 ghz,” IEEE Transactions on Wireless
ing (WCSP), Oct 2010, pp. 1–5. Communications, vol. 3, no. 3, pp. 922–928, May 2004.
[16] M. F. Iskander and Z. Yun, “Propagation prediction models for wireless [34] A. Durantini and D. Cassioli, “A multi-wall path loss model for indoor
communication systems,” IEEE Transactions on Microwave Theory and uwb propagation,” in 2005 IEEE 61st Vehicular Technology Conference,
Techniques, vol. 50, no. 3, pp. 662–673, Mar 2002. vol. 1, May 2005, pp. 30–34 Vol. 1.
[17] H. Okamoto, K. Kitao, and S. Ichitsubo, “Outdoor-to-indoor propagation [35] X. Liu, J. Cao, S. Tang, J. Wen, and P. Guo, “Contactless respiration
loss prediction in 800-mhz to 8-ghz band for an urban area,” IEEE monitoring via off-the-shelf WiFi devices,” IEEE Transactions on Mo-
Transactions on Vehicular Technology, vol. 58, no. 3, pp. 1059–1067, bile Computing, vol. 15, no. 10, pp. 2466–2479, oct 2016.
March 2009. [36] X. Wang, L. Gao, S. Mao, and S. Pandey, “CSI-based fingerprinting for
[18] M. F. Khan and B. Wang, “Effective placement of femtocell base stations indoor localization: A deep learning approach,” IEEE Transactions on
in commercial buildings,” in 2014 Sixth International Conference on Vehicular Technology, pp. 1–1, 2016.
Ubiquitous and Future Networks (ICUFN), July 2014, pp. 176–180. [37] L. Gui, M. Yang, H. Yu, J. Li, F. Shu, and F. Xiao, “A cramer–rao
[19] S. Aust, R. V. Prasad, and I. G. M. M. Niemegeers, “Outdoor long-range lower bound of CSI-based indoor localization,” IEEE Transactions on
wlans: A lesson for ieee 802.11ah,” IEEE Communications Surveys Vehicular Technology, vol. 67, no. 3, pp. 2814–2818, mar 2018.
Tutorials, vol. 17, no. 3, pp. 1761–1775, thirdquarter 2015. [38] L. Li, W. Yang, M. Z. A. Bhuiyan, and G. Wang, “Unsupervised learning
[20] L. Liechty, E. Reifsnider, and G. Durgin, “Developing the best 2.4 ghz of indoor localization based on received signal strength,” Wireless
propagation model from active network measurements,” in 2007 IEEE Communications and Mobile Computing, vol. 16, no. 15, pp. 2225–
66th Vehicular Technology Conference, Sept 2007, pp. 894–896. 2237, may 2016.
[21] H. Sun, P. Yang, L. Zu, and Q. Xu, “A far field sound source localization
system for rescue robot,” in 2011 International Conference on Control,
Automation and Systems Engineering (CASE). IEEE, jul 2011.
[22] N. Doulamis, P. Agrafiotis, G. Athanasiou, and A. Amditis, “Human
object detection using very low resolution thermal cameras for urban
search and rescue,” in Proceedings of the 10th International Conference

633