NETWORK EMBEDDED

APPLICATION

ASSIGNMENT 1

NAME : SOMNATH KHAMARU

ID : 2020MT12098
([email protected])
Problem Statement
A networked embedded system has to be designed for a township of interest(approximately
300acres).
The township has an industrial manufacturing unit (mainly factory floors with a single floor,
i.e.ground floor) and residential apartments. Network architecture has to be designed to support the
following
1) Energy management in the residential areas of the township. The networked embedded system
should be designed to monitor the electronic appliances (consumer appliances), presence of human
beings inside the room, ambient temperature, and humidity inside the rooms. The data collected
from the residential area at a collection point (a central controller) will be used to monitor the
energy consumption of the building periodically, then device methods to reduce energy
consumption at each residence and also control the devices(turn off/on/regulate (eg: set ac
temperature accordingly)) in the buildings to reduce the energy consumption without causing
discomfort to the users.It may be noted that the control of devices is done based on several factors
in addition to the presence or absence of a person in the room. Other factors such as the history of
presence or absence of users, climatic conditions, etc will be used to decide on control so that an
optimal decision can be made. There should not be any restriction on adding or removing new
devices into the system. Each device is not directly reachable to the central controller. Both periodic,
as well as event based data (when energy consumption goes above a certain threshold), have to be
sent to the central controller. In the locality, there are two types of residential buildings-

Category 1: Buildings that are under construction and hence can be designed as smart buildings.
Appropriate wired or wireless communications can be incorporated in such buildings and Category
2: the buildings whose construction is complete and can be “retrofit”to make them“smart”. Only
wireless communication can be used here. And the range of communication ion cannot be more
than 20m. The network design should be scalable (to be used for more number of buildings and
devices), however for demonstration,a prototype has to be implemented in the following
locality.Characteristics of smart buildings(Category 1)They are buildings with single levels. Such 10
buildings are to be monitored. Each building has 7 rooms with a minimum of 2-7 appliances to be
monitored in each room.Characteristics of buildings(Category 2)These are 5 apartments are to be
monitored. The apartment has 12 levels, each level with 5 houses.Each house has 7 rooms. Each
room has 3 devices whose energy consumption is to be monitored. There is a home level controller,
which aggregates data from the individual devices in the house and there is also a controller/level
(level controller) of the apartment. In addition to this,there's an apartment level controller, which
will aggregate data from the level controller. The apartment level controller communicates with a
central controller which monitors the energy consumption of smart buildings.Control messages can
also be sent from the central controller to the appliances so that they can be controlled accordingly(
You may assume that control messages can be sent at any point in time).
2) Condition-based monitoring of types of equipment in the industrial manufacturing unit(on the
factory floor).The company’s product is gearboxes for the automotive industry, with a production
volume of approximately 150,000 units per year. CBM is done as online monitoring(continuous
monitoring has to be done at the central controller (same as discussed for energy monitoring). One
of the most common problems of rotating equipment such as spindles (used in the manufacturing of
gearboxes) is the bearing condition (due to the wear of the bearings). Failure of the bearings can
cause major damage to a spindle. Vibration analysis can diagnose bearing failures by measuring the
overall vibration of a spindle or, more precisely, by frequency analysis is. Suitable Inertial
measurement units can be used for the recording of vibrations. Currently,there are 50 such units in
the company but more such units can be later included.
CBM is being used to plan maintenance action based on the condition of the machines and to
prevent
failures by solving the problems in advance as well as controlling the accuracy of the machining
operat ions.The spindle system is one of the critical subsystems of a machine tool that supplies the
necessary power to the cutting process. The spindle is a high-precision component that comprises
several parts, e.g., the rotor shaft, bearings, and the clamping system. The key components of the
spindle can be listed
as follows: 1) the spindle design, i.e., belt-driven or with an integral motor; 2) the bearing design,
including the type, tolerance, and method of lubrication; 3) the motor design, which comprises the
belt type, motor spindle, capacity, and kW rating; 4) the spindle shaft, which can be a tool retention
drawback, and the tooling system used; and 5) the spindle being used, which comprises the size,
mounting style, and capacity. All these components need to be in a perfect balance to achieve the
required high accuracy at elevated rotational speed and high material removal rate under stable
conditions. Like other rotating equipment, the most critical part of a spindle unit is the bearing.
Vibration monitoring can be a solution to predict spindle bearing failure and prevent long production
stops. 3) Monitoring of assets on the factory floor.A wireless tag can be mounted on the asset and
the same can be tracked at the central controller in real-time.
Design the network architecture for the same highlighting the following aspects.Note that only one
integrated system has to be designed (An industrial network and ahom e automation network at a
are collected at a common central controller-which is cloud-based). Data processing at the controller
is not required to be done. However, you may assume that suitable control messages are generated
based on the data collected.
a. Describe the type of nodes in the system-Mention the sensors and communication mechanism in
each type of node used in the system.
b. Describe the deployment of nodes in the system Mention assumptions such as type of sensors
used, the distance between placement of nodes,availability of line power,etc.You may use diagrams.
c. Explain the type of traffic(eg: periodic, query,control, data,etc)required for this application.
d. Mention the major design considerations for such a network?Analyze the robustness of the
system. How will you make the system robust and fault-tolerant?What are the quantitative
parameters that you will use for the performance analysis of the system?
e. Mention reasons, if wired or wireless communication is used for the system (Note that a
combination of the wireless or wired system may be used as and when required).
f. Mention the method of addressing nodes.
g. Mention the complete protocol stack for the network. Describe if time synchronization is
required. Which node will be the master node here? Describe the synchronization process.Is
localization required? Justify the answer. If required, suggest the method of localization Discuss the
MAC protocols required for the application. If the selected standard defines a MAC protocol (which
you will be utilizing) mentions the same.Discuss the routing protocol.Transport or higher layers
required?
h. Show the complete network architecture including the buses and gateways as and when required.
k. What are the security vulnerabilities you may have in this system?How will you ensure that system
is robust to those security vulnerabilities. Highlight The assumptions if any.

Solution
Home automation Network structure of Category 1
residential building
Design considerations for such a network:
 1. As the category 1 building are newly constructed so the motes placement is already
considered, and appropriate power sockets have been kept in desired location. Hence,
we will assume all motes will have wired power.
2. As the buildings are on of single level so they will have only home controller and
apartment-level controller and no level controller.
3. The network will be completely wireless based ZigBee network on IEEE standard
802.15.4 for low-rate wireless personal area networks (WPANs). Being a wireless
network adding new nodes to the system is easy in both operational and network
parameter. Only the Apartment level controller will have wired ethernet connection to
connect to the cloud via internet. As Apartment controller will be 1 per Apartment so
wired communication will helpful both in term of its capability to handle high data
rate and more reliability for edge to cloud connectivity.
4. The Apartment level controller will have ethernet based internet connection to connect
to the Central Controller
5. We will be using both mesh and Star topology in this network.

Types of Node
In the system mentioned in the problem statement for category 1 building (Buildings that are under
construction and hence can be designed as smart buildings. Appropriate wired or wireless
communications can be incorporated in such buildings), there can be 4 types of nodes Sensing Node
, Cluster Head , Gateway and Base Station.
Sensing Node – These nodes will have sensor embedded in them and can gather data using
those sensors and pass it on to the cluster head.
Cluster Head Node: These are mainly the home controllers , level controllers , which mainly
aggregate the data and pass it to the next level cluster head.
Gateway Node: The apartment controllers are mainly the gateway , which to are connected
to the cloud based controller and pass the aggregates data to Cloud for further processing
and decision making.
Base Station: The cloud based central controller, which has the highest computational
power in the network and has the best processing capability.

Sensing Node
The Sensing nodes will use ZigBee as the communication medium is non beacon enabled
mode , hence ZigBee router and coordinators will always be awake . There will be 4 types of
motes
Type 1 mote – Motes containing temperature and humidity sensor , these will be used to
captured ambient temperature and humidity in the room. The mote will talk to the sensor
using SPI/I2C protocol. These motes will mostly sleep and occasionally wake up to send data
to the ZigBee routers.

Type 2 mote – These Motes will be having Bidirectional IR sensors to detect people entry
and exit in a room and thus help detect the presence of no of humans in the room. The mote
will talk to the sensor using serial communication. These motes will be always on.
 Type 3 mode – These motes will be used to sense the power consumed in different
appliances . They will be able to drive high current devices 5A -15 A range . They will be
fitted to AC , TV and other electrical appliances. The sensors will communicate to the motes
via serial communication. These motes will mostly sleep and occasionally wake up to send
data to the ZigBee routers.

Type 4 mode – These motes will be used as ZigBee router (ZR), they will be used to pass data
from sensor to the Cluster Head. Their main task is relaying and is used to increase the
coverage of the network. These motes will be always on. Although they do no do any
sensing but they are considered as special sensing node as they do not aggregate the data
either.

Cluster Head Node

The Cluster Head Nodes will use ZigBee as the communication medium is non beacon
enabled mode , hence motes will be awake all the time as they have wired power , so no
sleeping is required to save power . This node will be the home controller . These motes will
be the ZigBee Coordinator. They will aggregate the data received from the ZigBee Router
and send it to Apartment controller.

Gateway Node
The Gateway node will use dual communication system . These mote will have dual role ,
They will acts as ZigBee controller(ZC) and get data from other ZC ( home controller). On the
other hand they will connect to cloud based central controller , hence they will have
ethernet based internet connection.

Deployment pattern of the Motes

CATEGORY 1 HOUSE MOTES DEPLOYMENT
Type of Traffic
o Unicast and broadcast mode of traffic is expected in the network when using ZigBee
protocol.
 o In Unicast mode the Sensing node will send periodic data to ZigBee controller( Home
controller ) .
o ZigBee routers transmit periodic beacons to confirm their presence to other network
nodes
o Broadcast traffic is sent when data commands are sent from central controller to
sensing nodes . The broadcast messages are send to all the nodes to find the route to
the specific sensing node.

Fault Tolerance and Robustness of the system

In this system we will used ZigBee Acknowledgement based Reliable Broadcast (ZARB) protocol . It
utilizes the acknowledgment mechanism in multicast and broadcast transmission for reliable
transmission by smartly managing the broadcasted acknowledgments for reliable data broadcast in
WSN.

As we are using Mesh topology which supports multi-hop communication (ZigBee End Device < ------
> ZigBee Router < ------- > ZigBee Coordinator) , through which data is passed by hopping from node
to node using most reliable communication link and most cost- effective path until the destination
node has been reached. The multi-hop ability also adds to the fault tolerance, if one node fails or
experience radio interference, the network will reroute itself using the remaining nodes.

Quantitative parameters used for performance analysis of the

system

 The mean delay of the message packet. It should be under permissible limits
 The packet discard rate . It should also be under permissible limit
 Availability of the node . Nodes should have high uptime so the throughput of the system is
maintained.
 Power usage by the node network.

Method for addressing the nodes

A mote receives a 16-bit network address when it joins the ZigBee network. The 16-bit address of
0x0000 is reserved for the ZigBee coordinator mote. All other motes will receive a randomly
generated address from the ZigBee router mote or ZigBee coordinator mote that allows the join. The
16-bit address can change under certain conditions:

 An address conflict is detected where two motes are found to have the same 16-bit address
 A mote leaves the network and later joins (it can receive a different address)

All ZigBee transmissions will be sent using the source and destination 16-bit addresses. The routing
tables on ZigBee motes also use 16-bit addresses to determine how to route data packets through
the network. However, since the 16-bit address is not static, it is not a reliable way to identify a
mote.

To solve this problem, the 64-bit destination address ( MAC address of the mote) can be included in
data transmissions to guarantee data is delivered to the correct destination. The ZigBee stack can
discover the 16-bit address, if unknown, before transmitting data to a remote.
Time Synchronization
I chose to implement the Flooding Time Synchronization Protocol (FTSP). This protocol was
developed to
demonstrate a means of providing network-wide time synchronization to a large network of wireless
sensors. The FTSP was designed to be robust despite the failure of individual nodes (a necessary
consideration in a WSN like this). The FTSP generates time synchronization with periodic time sync
messages. The FTSP can provide a WSN with high-precision synchronization without requiring a lot
of resource overhead.
The FTSP can be optimized for a master-slave network ( as we are using a ZigBee start topology) such
that only the master nodes ( ZigBee router) transmit the periodic time sync messages and the slave
nodes ( sensing motes)
simply receive without having to transmit at all. A slave node will even disable its radio and enter
low-power mode between messages. And highly accurate time synchronization like FTSP could
enable a slave node to sleep for a very long time and still wake up just before the next message was
due, making the most of the power conservation strategy.

Localization
Each of the motes are identified by their network address. In this building home automation setup ,
none of the motes will be moving and even if they are moved then it will not hamper the network.

MAC protocol
The data transmission mechanism in ZigBee networks is supported by two operational modes: with
or without beacon enabled mode. The network will use CSMA MAC protocols. When sensing mote
wishes to transfer data in a non-beacon enabled mode, it simply transmits its data frame, using
unslotted CSMA/CA, to the coordinator. The coordinator will acknowledge the successful reception
of the data by transmitting an optional acknowledgment frame. In non-beacon enabled mode, the
generation of periodic beacon frames is disabled. So, devices simply transmit data frames using
unslotted CSMA/CA to the coordinator. The PAN coordinator is responsible for handling only mote
association/disassociation. The advantage of the non-beacon enabled mode, with regard to WSN
application requirements, is that it easily allows scalability and self-organization. However, the non-
beacon enabled mode does not provide any guarantee to deliver data frames, specifically within a
certain deadline.

When a sensing mote wishes to transfer data to a coordinator in a beacon enabled mode, it will first
listens for the network beacon. When the beacon is found, the mote synchronizes with the
superframe. At the appropriate time, the mote transmits its data frame, using slotted CSMA/CA, to
the coordinator. The coordinator may acknowledge the successful reception of the data by
transmitting an optional acknowledgment frame. Beacon enabled mode makes the IEEE 802.15.4
protocol sufficiently flexible and able to perform the needs of a large variety of applications. It has
the ability to provide guaranteed real time performance to the network. The beacon enabled mode
uses beacons which are periodically sent by the PAN coordinator to identify its PAN and synchronize
nodes that are associated with it. The beaconing synchronization mechanism in IEEE 802.15.4
enables networks to work under a controllable duty cycle to achieve better energy efficiency
compared to the non-beacon enabled mode.

Routing
ZigBee network employs mesh routing to establish a route between the source mote and the
destination mote. Mesh routing allows data packets to traverse multiple nodes (hops) in a network
to route data from a source to a destination. Routers and coordinators can participate in establishing
routes between source and destination devices using a process called route discovery. The routing
protocol used by our ZigBee the network layer is ZigBee-Ad-hoc On-demand Distance Vector routing
(Z-AODV) protocol. To find a destination mote, Z-AODV is used to broadcast a route request to all of
its neighbours. The neighbours then broadcast the request to their neighbours and onward until the
destination is reached. Once the destination is reached, a route reply is sent via unicast transmission
following the lowest cost path back to the source. Once the source receives the reply, it updates its
routing table with the destination address of the next hop in the path and the associated path cost.
Z-AODV is quite suitable of mesh routing as in case of our ZigBee network.

Home automation Network structure of Category 2

residential building
Types of Node
In the system mentioned in the problem statement for category 2 building (buildings whose
construction is complete and can be “retrofit”to make them“smart”. Only wireless communication
can be used here. And the range of communication cannot be more than 20m.), there can be 4 types
of nodes Sensing Node , Cluster Head , Gateway and Base Station
Sensing Node – These nodes will have sensor embedded in them and can gather data using
those sensors and pass it on to the cluster head.
Cluster Head Node: These are mainly the home controllers , level controllers , which mainly
aggregate the data and pass it to the next level cluster head.
Gateway Node: The apartment controllers are mainly the gateway , which to are connected
to the cloud based controller and pass the aggregates data to Cloud for further processing
and decision making.
Base Station: The cloud based central controller, which has the highest computational
power in the network and has the best processing capability.

Design considerations for such a network:

1. As the category 2 building are old and construction is already completed , so most of the
motes will be having battery power as their placement were not considered during the
building design ,hence appropriate power sockets will not be present. Nodes used to
track power consumption of electrical appliances , will have access to wired power.
2. The network will be completely wireless ( as the part of the requirement itself) based
ZigBee network on IEEE standard 802.15.4 for low-rate wireless personal area
networks (WPANs). Being a wireless network adding new nodes to the system is easy
in both operational and network parameter. Only the Apartment level controller will
have wired ethernet connection to connect to the cloud via internet. As Apartment
controller will be 1 per Apartment so wired communication will helpful both in term of
its capability to handle high data rate and more reliability for edge to cloud
connectivity.
3. The Apartment level controller will have ethernet based internet connection to connect
to the Central Controller.
4. We will be using both mesh and Star topology in this network.
5. Each room has only one door .

Sensing Node
The Sensing nodes will use ZigBee as the communication medium is non beacon enabled
mode , hence ZigBee router and coordinators will always be awake . The ZigBee Routers and
Coordinator will have wired powered. There will be 4 types of motes
Type 1 mote – Motes containing temperature and humidity sensor , these will be used to
captured ambient temperature and humidity in the room. The mote will talk to the sensor
using SPI/I2C protocol. These motes will mostly sleep and occasionally wake up to send data
to the ZigBee routers. These motes will run on battery.

Type 2 mote – These Motes will be having Bidirectional IR sensors to detect people entry
and exit in a room and thus help detect the presence of no of humans in the room. The mote
will talk to the sensor using serial communication. These motes will be always on but run on
battery. These motes will keep the radio off most of the time as they don’t need to hear to
any polling and would be added via start topology. They will do the sensing all the time and
once some entry or exit of human is sent then , it will switch on the radio to send that info to
the router and again go back to radio sleep mode , thus power would be saved.

Type 3 mode – These motes will be used to sense the power consumed in different
appliances . They will be able to drive high current devices 5A -15 A range . They will be
fitted to AC , TV and other electrical appliances. The sensors will communicate to the motes
via serial communication. These motes will mostly sleep and occasionally wake up to send
data to the ZigBee routers.

Type 4 mode – These motes will be used as ZigBee router (ZR), they will be used to pass data
from sensor to the Cluster Head. Their main task is relaying and is used to increase the
coverage of the network. These motes will be always on. Although they do no do any
sensing but they are considered as special sensing node as they do not aggregate the data
either.

Cluster Head Node

The Cluster Head Nodes will use ZigBee as the communication medium is non beacon
enabled mode , hence motes will be awake all the time as they have wired power , so no
sleeping is required to save power . This node will be the home controller , level controller.
These motes will be the ZigBee Coordinator. They will aggregate the data received from the
ZigBee Router and send it to Apartment controller ( home controller < --- > Level Controller <
--- > Apartment Controller).

Gateway Node
The Gateway node will use dual communication system . These mote will have dual role ,
They will acts as ZigBee controller(ZC) and get data from other ZC ( home controller). On the
other hand they will connect to cloud based central controller , hence they will have
ethernet based internet connection.
Deployment pattern of the Motes
 Each Room will have one Type 1 node mostly at the center of the room and one type 2 node
at the door.
 Type 3 node per room will depend on the no of appliances monitored per room. Type 3
motes will be attached on each appliance to be monitored
 Type 4 node will place near to rooms where more no of type 3 nodes are kept as that room
will have more data traffic.

CATEGORY 2 HOUSE MOTES DEPLOYMENT

 CATEGORY 2 BUILDING HOME AUTOMATION NETWORK

Type of Traffic
o Unicast and broadcast mode of traffic is expected in the network when using ZigBee
protocol.
o In Unicast mode the Sensing node will send periodic data to ZigBee controller( Home
controller ) .
 o ZigBee routers transmit periodic beacons to confirm their presence to other network
nodes
o Broadcast traffic is sent when data commands are sent from central controller to
sensing nodes . The broadcast messages are send to all the nodes to find the route to
the specific sensing node.

Fault Tolerance and Robustness of the system

In this system we will used ZigBee Acknowledgement based Reliable Broadcast (ZARB) protocol . It
utilizes the acknowledgment mechanism in multicast and broadcast transmission for reliable
transmission by smartly managing the broadcasted acknowledgments for reliable data broadcast in
WSN.

As we are using Mesh topology which supports multi-hop communication (ZigBee End Device < ------
> ZigBee Router < ------- > ZigBee Coordinator) , through which data is passed by hopping from node
to node using most reliable communication link and most cost- effective path until the destination
node has been reached. The multi-hop ability also adds to the fault tolerance, if one node fails or
experience radio interference, the network will reroute itself using the remaining nodes.

Quantitative parameters used for performance analysis of the

system

 The mean delay of the message packet. It should be under permissible limits
 The packet discard rate . It should also be under permissible limit
 Availability of the node . Nodes should have high uptime so the throughput of the system is
maintained.
 Power usage by the node network.
 Battery life time of the nodes . If the network is properly optimised then better battery life
will be present.

Method for addressing the nodes

A mote receives a 16-bit network address when it joins the ZigBee network. The 16-bit address of
0x0000 is reserved for the ZigBee coordinator mote. All other motes will receive a randomly
generated address from the ZigBee router mote or ZigBee coordinator mote that allows the join. The
16-bit address can change under certain conditions:

 An address conflict is detected where two motes are found to have the same 16-bit address
 A mote leaves the network and later joins (it can receive a different address)

All ZigBee transmissions will be sent using the source and destination 16-bit addresses. The routing
tables on ZigBee motes also use 16-bit addresses to determine how to route data packets through
the network. However, since the 16-bit address is not static, it is not a reliable way to identify a
mote.

To solve this problem, the 64-bit destination address (MAC address of the mote) can be included in
data transmissions to guarantee data is delivered to the correct destination. The ZigBee stack can
discover the 16-bit address, if unknown, before transmitting data to a remote.
Time Synchronization
I chose to implement the Flooding Time Synchronization Protocol (FTSP). This protocol was
developed to
demonstrate a means of providing network-wide time synchronization to a large network of wireless
sensors. The FTSP was designed to be robust despite the failure of individual nodes (a necessary
consideration in a WSN like this). The FTSP generates time synchronization with periodic time sync
messages. The FTSP can provide a WSN with high-precision synchronization without requiring a lot
of resource overhead.
The FTSP can be optimized for a master-slave network ( as we are using a ZigBee start topology) such
that only the master nodes ( ZigBee router) transmit the periodic time sync messages and the slave
nodes ( sensing motes)
simply receive without having to transmit at all. A slave node will even disable its radio and enter
low-power mode between messages. And highly accurate time synchronization like FTSP could
enable a slave node to sleep for a very long time and still wake up just before the next message was
due, making the most of the power conservation strategy.

Localization
Each of the motes are identified by their network address. In this building home automation setup ,
none of the motes will be moving and even if they are moved then it will not hamper the network.

MAC protocol
The data transmission mechanism in ZigBee networks is supported by two operational modes: with
or without beacon enabled mode. The network will use CSMA MAC protocols. When sensing mote
wishes to transfer data in a non-beacon enabled mode, it simply transmits its data frame, using
unslotted CSMA/CA, to the coordinator. The coordinator will acknowledge the successful reception
of the data by transmitting an optional acknowledgment frame. In non-beacon enabled mode, the
generation of periodic beacon frames is disabled. So, devices simply transmit data frames using
unslotted CSMA/CA to the coordinator. The PAN coordinator is responsible for handling only mote
association/disassociation. The advantage of the non-beacon enabled mode, with regard to WSN
application requirements, is that it easily allows scalability and self-organization. However, the non-
beacon enabled mode does not provide any guarantee to deliver data frames, specifically within a
certain deadline.

When a sensing mote wishes to transfer data to a coordinator in a beacon enabled mode, it will first
listens for the network beacon. When the beacon is found, the mote synchronizes with the
superframe. At the appropriate time, the mote transmits its data frame, using slotted CSMA/CA, to
the coordinator. The coordinator may acknowledge the successful reception of the data by
transmitting an optional acknowledgment frame. Beacon enabled mode makes the IEEE 802.15.4
protocol sufficiently flexible and able to perform the needs of a large variety of applications. It has
the ability to provide guaranteed real time performance to the network. The beacon enabled mode
uses beacons which are periodically sent by the PAN coordinator to identify its PAN and synchronize
nodes that are associated with it. The beaconing synchronization mechanism in IEEE 802.15.4
enables networks to work under a controllable duty cycle to achieve better energy efficiency
compared to the non-beacon enabled mode.

Routing
ZigBee network employs mesh routing to establish a route between the source mote and the
destination mote. Mesh routing allows data packets to traverse multiple nodes (hops) in a network
to route data from a source to a destination. Routers and coordinators can participate in establishing
routes between source and destination devices using a process called route discovery. The routing
protocol used by our ZigBee the network layer is ZigBee-Ad-hoc On-demand Distance Vector routing
(Z-AODV) protocol. To find a destination mote, Z-AODV is used to broadcast a route request to all of
its neighbours. The neighbours then broadcast the request to their neighbours and onward until the
destination is reached. Once the destination is reached, a route reply is sent via unicast transmission
following the lowest cost path back to the source. Once the source receives the reply, it updates its
routing table with the destination address of the next hop in the path and the associated path cost.
Z-AODV is quite suitable of mesh routing as in case of our ZigBee network.

Industrial Network structure of the Factory Floor

Types of Node
In the system mentioned in the problem statement for the factory where condition based
monitoring is used.
Design considerations for such a network:
1. As the factory floor has industrial sensor which require high data rate and contain
wireless tags which very low data rate requirement, we will use a hybrid industrial
network wired and wireless combined.
2. As the main requirement in the factory is to do vibration monitoring of the bearing and
beaning is present in a rotating system , so we can expect high data rate so we will used
wired motes to collect the vibration data.
3. As the equipment’s will have a wireless tag so for them we will use wireless
communication to get their position.
4. The vibration pattern will be analysed in the central controller with predefined models
and any anomaly in the vibration behaviour will be detected by it.

Sensing node (Vibration sensor)

 The motes used in sensing vibration will have wired communication

 The vibration sensor data from the different IMU unit is carried by Profibus protocol .
 The PROFIBUS DP is used to operate sensors and actuators via a centralised controller in
production (factory) automation applications.
 PROFIBUS DP runs at speeds between 9.6 kbit/s and 12 Mbit/s. A particular speed can be
chosen for a network to give enough time for communication with all the devices present in
the network. The RS485 balanced transmission used in PROFIBUS DP only allows 31 devices
to be connected at once; But as we have 50 machine units in the company and many more
are to be added , so the network expanded with the use of hubs or repeaters (2 hubs or
repeaters to reach 126 ( max no of device using PROFIBUS).
 Sensing node (Wireless tag)

 The wireless tag will be battery operated and ble based .

 It will communicated periodically, say twice a day to the Bluetooth enabled gateway
mode .
 The position of gateway nodes will be static and preplaced and the relative position of the
tag will be determined by the gateway it is connecting.
 Tags will broadcast message about their presence and multiple gateway can pickup the
message and send the data to the IO controller of the factory.

SCADA/IO controller

 The data coming from the sensors is given to the IO controllers like SCADA. In the industrial
environment SCADA is the most robust and cost-efficient controller solution .
Sensor bus (PROFIBUS)

 PROFIBUS is used to operate sensors and actuators via a centralised controller in production
(factory) automation applications.
 PROFIBUS it is based on a master–slave scheme which implements data exchanges between
controllers (“masters”) and field devices such as sensors and actuators (“slaves”) at the
maximum speed of 12 Mb/s. This is suitable for high-speed rotational system like us.
 The data-link layer protocol of Profibus DP, known as fieldbus data link (FDL), is based on a
token-passing scheme. A special frame called token, which grants exclusive access to
  the physical medium is continuously circulated among the master stations during the
network operation.
 Slave devices are only allowed to access the network (i.e., to transmit data) when they are
queried by a master station.
 The operation of Profibus DP relies on a polling cycle that is repeated continuously by the
master on slave nodes. During the cyclic operation, the master sends a request telegram to
each slave containing the output data. At the same time, it fetches input data made
available from the queried device, which are included in the response telegram.

Field Bus (PROFINET for plant automation)

 For Field bus Profinet is used .Profinet is an industry technical standard for data
communication over Industrial Ethernet, designed for collecting data from, and controlling
equipment in industrial systems, with a particular strength in delivering data under tight time
constraints.
 The application layer protocol of Profinet IO in based on the definition of communication objects that
can be exchanged over the network via communication relationships established between I/O
controllers and I/O devices.
 The medium access protocol of Profinet IO is based on a time division multiple access (TDMA)
technique. In practice, the traffic is scheduled according to a cycle, which consists of four phases, as
shown in the picture below.

Deployment pattern of the Motes

The Vibration sensor will be placed on each spindle and the mote will be connected via profibus

Each equipment will have wireless tag . The wireless tag will have 10 meter range , as the range is
less the battery on the tag will last of longer duration. The Bluetooth gateways would be kept 20
meter apart so that the more or less the tag always resides in between two gateway nodes. The
factory floor will be divide in to circles of 20 meter radius to place the gateways and the placement
will be on the celling of the factory to avoid any obstruction in the factory floor
Type of Traffic
 Periodic information will be sent continuously to the central controller for analysis. Control
data packets are will be mostly used.
 As all the salve sensors data will collected by the master and will be sending the data to
central controller a converge cast kind of traffic is expected at the central controller.
 Periodic traffic will be sent by the wireless tags present on the equipment to mention its
presence.

Fault Tolerance and Robustness of the system

 We will be using pluggable nodes that can be removed/replaced without impinging
on network operation.
 Appropriate network layout and segmentation is done so that physical layer faults
allow critical plant operation to continue in the event of failure or node replacement.
 For PROFIBUS systems we will use Connector systems and layouts that do not break
the bus or loose termination when disconnected. Also use termination solutions that
allow devices to be removed or replaced.
 Use appropriate solutions for redundancy.

Quantitative parameters used for performance analysis of the

system

 The mean delay of the message packet. It should be under permissible limits
 The packet discard rate . It should also be under permissible limit
  Availability of the node . Nodes should have high uptime so the throughput of the system is
maintained.
 Power usage by the node network.
 Battery life time of the nodes . If the network is properly optimised then better battery life
will be present.

Method for addressing the nodes

Each vibration sensors used in the network will be uniquely identified by the mac address and the
sensor id. Each slave node belonging to same mac address will be identified with the unique
vibration sensor id only.

The wireless tags will be identified by their unique MAC id.

Time Synchronization
The Network Time Protocol, or NTP is used for Time Synchronization. As the network is connected
to internet and hence the TCP/IP synchronizes all the layers of the protocol with the global clock
setting derived from the internet. The data packet can be time stamped at the central controller
level.

Localization
As the vibration sensors are placed at fixed location on the fixed machine an explicit localization
technique is not required.

Localization is required for wireless tags present on equipment. When the wireless tags broadcast
their presence to the gateways, then multiple gateways can pickup the signal and send the data
along with the RSSI to master node via the industrial bus. The central controller on receiving the
data, check the RSSI mentioned in the packet and compares the RSSI of all the packets of a particular
wireless tag and used triangulation algorithm to find the location of the tag.

MAC protocol
The MAC protocol Profinet IO is based on a time division multiple access (TDMA) technique. In
practice, the traffic is scheduled according to a cycle, which consists of four phases .Thefirst phase
(RED) is reserved for RT_CLASS 3 traffic (the most critical) that takes place over predefined physical
links enforced by a special kind of switches purposely developed or Profinet IO. In the ORANGE
phase, only RT_CLASS 2 frames are exchanged. This is still time critical traffic but requirements for
physical links are not defined. The GREEN phase is reserved for both RT_CLASS 2 and RT_CLASS
frames, either cyclic or acyclic. This is the only mandatory phase. The access to the physical medium
is regulated by the priority assigned to frames. Finally, the YELLOW phase is reserved for non-real-
time traffic. Yellow phase is used for traffic from wireless tags .

Routing

The Routing of the sensor data is managed as per PROFIBUS /PROFINET protocol.
Security vulnerabilities
 Threat of Human intervention. Some person can tamper the system . May remove power
from Controller .
 Gateway connecting to the Central Controller may be hacked ( as it is connected to internet)
and the Central controlled might get flooded by wrong messages and creating a DDOS
attack.
 As we are using ADOV routing in the home automation network so the network might be
affect by a Rushing Attack , which on a sensor network exploits the nature of the route
discovery procedure of on-demand routing protocols like ADOV. In this type of attack, a
malicious node immediately forwards incoming route request messages to its neighbours,
therefore “rushing” these messages without consideration of any protocol. Therefore, the
node has an increased probability of being part of the chosen route between source and
destination.
 Attack on Data aggregation System can happen. Even simple aggregation functions can
easily be influenced by an attacker such that a network’s behaviour can be altered .

System Robustness against the above-mentioned security

vulnerabilities
 DDOS attack from the gateway can be tackled by the central controller by removing the
affected gateway ( say apartment level controller ) and re-route the traffic from other apart
controller. We assume each apartment will be close enough so that every apartment
controller is ZigBee communication range of another two-apartment controller.
 Attack on Data Aggregation system like home controller and ZigBee router can be curbed
by adding in delays both aggregation and authentication, for example, aggregation does not
take place at the first hop that would be able to perform this aggregation, but at the second
hop. While this may increase resource expenditures and system pace, it may also enable
integrity guarantees where consecutive nodes have not been compromised.
 Rushing Attack in the ZigBee network can be curbed using combination of several protective
measures can be used. A secure neighbour detection approach can be used to allow both
the sender and the receiver of a route request to verify that the other party is in fact within
the normal transmission range to prevent attacker from far off place to send high energy
radio transmission to forward route request.

References
1. Embedded System Handbook , Richard Zurawski
2. FUNDAMENTALS OF WIRELESS SENSOR NETWORKS THEORY AND PRACTICE , Waltenegus
Dargie
3. https://www.digi.com/resources/documentation/Digidocs/90002002/Concepts/c_zb_addre
ssing.htm?TocPath=Transmission%2C%20addressing%2C%20and%20routing%7C_____1
4. https://www.hindawi.com/journals/jcnc/2015/697153/
5. Routing in ZigBee: benefits from exploiting the IEEE 802.15.4 association tree , Francesca
Cuomo, Sara Della Luna, Ugo Monaco University of Roma “La Sapienza”, INFOCOM Dept.
Via Eudossiana 18, 00184, Rome, Italy.
6. Localization in Zigbee-based Sensor Networks ,Ralf Grossmann, Jan Blumenthal, Frank
Golatowski, Dirk Timmermann ,CELISCA, Center for Life Science Automation,Friedrich-
Barnewitz-Str ,University of Rostock.
7. A Comparative Review on Reliability and Fault Tolerance Enhancement Protocols in Wireless
Sensor Networks , Abhinav Sharma1, Sandeep Sharma2 ,Guru Nanak Dev University,
Punjab, India
8. https://wisilica.com/company/top-6-iot-communication-protocols/