A Sophisticated Secured Smart Metering System

Mohamed Hassan Abdel Hamied Essam M. Ramzy Hamed Ayman M. Bahaa-Eldin

Arab Academy for Science and Arab Academy for Science and Misr International University
Technology and Maritime Transport Technology and Maritime Transport On Leave From Ain Shams University
College of Computing and Information College of Computing and Information Faculty of Engineering
Technology (CCIT) Technology (CCIT) Cairo, Egypt
Cairo, Egypt Cairo, Egypt [email protected]
[email protected] [email protected]

Abstract— Security has become very important for the smart without a need for supply from the macrogrid [3] using
metering system to maintain its user privacy and prevent its renewable energy sources (RES). The information subsystem
vulnerability which may cause information misuse and prevent is responsible for the metering data that is collected from
it from sabotaging the system. Cyber-attacks, if not prevented, customers, sensors and phasor measurement units (PMUs) to
may cause instability in the energy supply which in turn monitor the grid state [3]. The communication subsystem is
negatively affects the infrastructure of the grid. Previous responsible for connectivity and exchange of information. It
metering systems have many disadvantages such as the presence can be wired (ADSL, PLC, Fiber) and wireless (WIFI, LTE,
of transfer keys that have no encryption, the use of large Zigbee, WIMAX, Bluetooth) [5]. Smart management is the
numbers of keys in initial state, and the collection of
brain of the smart grid which uses a set of tools and a system
consumption readings which depends on the number of smart
meters in the chain. In this paper, an innovative authentication
software such as supervisor control and data acquisition
and security system designed for smart metering in smart grids (SCADA), remote terminal unit (RTU), meter data
is proposed. This new system is distinguished by its ability to management system (MDMS), outage management system
protect the two ways of communication and its use of a small set (OMS), load management system (LMS), and demand
of keys while not adding much complexity to the message size. response management system (DRMS) [6]. The smart
Combining routing security to protect the gateways and access protection is responsible for electricity reliability, recovery
control list that monitors and filters traffic to enhance the (self-healing) from failure state, and prevention of asset attack
overall security, the proposed model in this article proved to be and cyber-attack [1], [7].
efficient and secure.

Keywords— smart grid, smart metering system, bit-wise

operation, OSPF routing protocol, traffic filtering

I. INTRODUCTION
Modern life is mainly based on using electricity.
Therefore, enough electricity has to be generated for all types
of customers in different locations (home, factory, etc.) which
in turn leads to an increase in fossil fuel emissions threatening
life on Earth. People nowadays dream to live a smart life
enjoying smart homes that allow the dweller to automate
control of lights, temperature, sound system, and security
cameras. As a result, a need arises to develop a new
technology that meets these customers’ increasing and
changing demands. Classical electricity grid that is formed of
a set of assets and generates and transmits electricity in one
way from the generation site to the consumption site cannot
meet the new customers’ demands. Smart grid, as a result, is
introduced as a new model of electricity grid that allows
information and power flow on both ways between utility and
customers as shown in Fig. 1, and is classified into three Fig. 1. Smart grid electricity and information flow framework
technical perspectives: smart infrastructure, smart
management and smart protection [1], [2]. This paper is organized as the following: Section II
describes the smart metering infrastructure and attack
The smart infrastructure allows the flow of energy in both analysis. Related works are discussed in section III. Section
directions and consists of energy, information and IV describes the three levels of security in the proposed model.
communication subsystems [3]. The energy subsystem The experimental results found in section V, and the
contains generation, transmission and distribution [4]. The conclusion is in section VI.
generation level of electricity in smart grid shows a new
pattern developed to generate electricity in a distribution level II. SMART METERING INFRASTRUCTURE AND ATTACKS
called “distributed generation” (DG) using a distributed
energy resource (DER) like solar systems and wind turbine. In The smart metering is an important part of the smart grid.
the case of power outages from the electricity grid, the It is a new technology that will provide a new model for the
distributed generation will provide electricity to the microgrid measurement system and is completely different from the old

978-1-7281-3780-3/19/$31.00 2019
c IEEE 171
 system. It provides customers with information on how to use consumption and enables the remote control in the connection
and rationalize energy consumption. It consists of, Smart and separation of electricity. As a result, it keeps the privacy
meter: is a smart device installed in the customer's site to of customers and secures the data traded between utility and
measure the consumption of electricity used by the customer customers from people with bad intentions.
in real time. It communicates by a wireless or wired network x Attack Impetus: There are many purposes that
with utility to transfer data for billing and management empower malicious users to attack a smart metering
purposes [8]-[9]. Advanced metering infrastructure (AMI): it system such as theft: Monitoring specific meters, as a
is an integrated system that contains smart meters, result, became possible through analyzing the
applications, communication devices to (send / receive) data electricity consumption behavior to know the time
between the smart meter and utility such as the ability to people are at home. Thus, facilitating theft as shown in
automatically and remotely measure electricity use, connect Fig. 3, Avenging: If an employee was fired, he would
and disconnect service, identify and isolate outages, and try to take revenge by decreasing the utility earning via
monitor voltage [10]-[12]. causing failures, Challenge: Attackers with bad
intentions may invade the systems of big companies,
A. Components of Smart Metering and War and Terrorism: Electricity grid has become a
The components of the advanced metering infrastructure national security. A country's enemies or terrorists can
[13] as shown in Fig. 2 are: get important information to do blackout to hurt a
victim.
x Smart devices: These are the devices which the
customer has in the customer site and are connected
with the smart meter through the home energy
management system (HEMS) to control resources
inside the home area network (HAN) and provide a
demand response [14].
x Smart meters: These calculate the customer
consumption and send this information periodically to
the collector. It connects with the collector through the
neighbor area network (NAN).
x Collector: It collects consumption from different
meters and throws it to the distributed local
management through a Wide Area Network (WAN).
x Local management: Customer's consumption data is Fig. 3. Electricity consumption through one day
stored in MDMS to control the user billing and
operators using this data with other applications such x Attack Types: There are two types of attacks that might
as OMS, LMS/ DRMS, and customer information happen on communication in a smart grid, passive
system (CIS) to control the grid status. attacks and active attacks. In both types, attackers are
looking for weak points to perform their attacks on the
x User application: It is used by the customer’s smart smart metering [15]-[16]. Passive attacks: The attacker
phone to pay and monitor them consumption. monitors the data but does not do any modification to
the data or cause harm to the grid infrastructure such
as eavesdropping [7]. Active attacks: The attacker
monitors data and does modifications to the transferred
data such as (management message, automatic meter
reading) and harmful things to the grid infrastructure
such as (shutout any hardware device in smart
metering system). As denial of service, the attacker
makes massive requests to down any part of the grid
[17], Spoofing Intruder sends data to control center or
management message to smart meter to appear as if it
is coming from an authorized node [18].
C. Attacks Countermeasure Tools
It is expected that there are huge amounts of data, bigger
than the Internet capacity, that will be transferred over the
smart metering system. This is because the devices at the
customer's will be connected to a smart grid [8]. CIA model
Fig. 2. Advanced metering infrastructure framwork
will achieve the suitable security for smart metering
transmitted data [19]-[20].

B. Attack Analysis x Cryptography Services: Used to supply confidentiality

Many governments now started resorting to installing and authentication by Converting a readable message
millions of smart meters in customer sites and the number is to a non-readable message through which the receiver
increasing daily. This system allows utility to control will be the only one who has the ability to decipher the

172 Third World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4)
 text [3]. Classified into symmetric and asymmetric its own unique pattern key to communicate with another node
algorithm, Symmetric algorithm using the same key to in the network using RTS/CTS frame. But this algorithm has
convert readable message to unreadable message and several problems to be addressed 1) It allows for attaching the
vice versa, asymmetric algorithm using pair key one to key to RTS frame and transfer in the communication network
convert readable message to unreadable message and without any encryption. 2) It does not determine the schedule
other to vice versa [6], [21]-[22]. Hash function it is of consumption reading sent by smart meters. 3) It does not
one-way function take any length of input data and specify how to transfer the management messages.
using mathematical model to calculate fixed length of
hash, receiver doing the same hash calculation and Vijay Kumar Proposed bootstrap protocol [31] using 4
comparing with received to assure integrity [6], [21]. keys to authenticate wakeup smart meter and symmetric
session key to encrypt transferred data from smart meter to
x Routing protocol: is used to configure and reconfigure control center. The proposed protocol has several problems to
in recovery state after the failure of incidence inside be addressed as follows: 1) Using a large number of keys for
distribution network [23]. Classified into static and boots. 2) The author supposes that the smart meter is
dynamic [24]. Open Shortest Path First (OSPF) in the connected directly to the local management or collector but he
proposed model. It is an open standard link state does not determine how smart meters authenticated together
protocol developed to be used in a large company. It for communication purpose.
calculates the shortest path by using DIJKSTA
Ye Yan Proposed [32] integrated authentication and
algorithm and performs neighbor's relationship that is
confidentiality (IAC) protocol using neighbors authenticated
called “adjacency” by exchanging Hello Packet with
smart meter as an authenticator for new start up smart meter
the specific parameters such as hello interval, dead
with a four-way handshake. The proposed protocol has
interval, area flag and authentication. After that,
problem as follows: 1) Hop-by-hop for billing data and
information is shared with neighbors, while the Hello
management message between the smart meter and the
Packet is considered as a keep-alive tool to discover if
collector makes the size of the encrypted information large
neighbor's relationship is unavailable. The backbone
and equivalent to the chain.
router connects to area 0 and the area border router
(ABR) that connects at least two different areas [25]. Utku Ozgur proposed [33] developed system provides
OSPF uses authentication to secure routing update and privacy with the paillier cryptosystem and two authentication
prevents malicious or incorrect routing information. It factors openSSL and ECDSA. implement the aggregation
has two types of authentication clear text and model over the realistic testbed (Beaglebone Black Board) and
cryptographic using MD5 [26]. compare the results with the NS3 simulation-based solution.
The proposed protocol has problems as follows: 1) Commands
x Traffic filtering: Inexpensive way to isolate malicious messages are not simulated. 2) TCP parameters are different
and unnecessary traffic from accessing the network by between sim and testbed.
analyzing incoming and outcoming traffic based on a
set of parameters such as IP source and destination, Samet Tonyali [34] design multicast firmware update
port source and destination and protocol [27] used to protocol provides confidentiality, access control, and message
face denial of service attack that sends sufferer authentication. The proposed protocol has a problem as
plurality of requests to fill up victim resources such as follows: That the structure of access tree affects the size and
bandwidth, processor and memory to prevent the computation time of signcrypt/designcrypt process.
sufferer from responding to any new requests [28].
Pardeep Kumar [35] Provide two-way security scheme
x Bit-wise: In the proposed model, suppose that all AES that is secure and energy efficient (computational and
keys are built in main chip during manufacturing so we communication). The proposed protocol too has problem as
cannot use the same keys to authenticate nodes and follows: MAC encryption algorithm adds the burden of the
encrypt data forever. It uses the bit-wise technique to encryption process.
generate encryption result. The DNA is composed of
four types of nucleotide namely A, C, G, T (A - IV. PROPOSED MODEL
adenine, C - cytosine, G - guanine, T - thymine) [29]- The proposed model consists of three levels of security to
[30], so suppose all smart meters and Autonomous secure the AMI communication. The first level is the routing
system (AS) is also called service provider (SP) or security level, the second level is the traffic filtering (packets
local management center have reference tables for filter) and the third is the proposed End-to-End (E2E) level.
nucleotide. At the authentication step, AS generates
DNA sequence (S) that is composed of nucleotide A. Routing Security Protocol Level
letters. Smart meters and AS convert the consumption The objective of this level is to authenticate exchange
reading to binary and DNA sequence (S) according to routing messages to prevent rouge from injecting fake routing
reference table. Bit-wise will perform using one bit information and causing unstable network or DOS attack. The
from reading binary with one bit from DNA sequence proposed model using OSPF logical hierarchy architecture
(S) and finally convert result of this operation to string with two types of areas. First type: Is backbone area (area 0).
then complete the model steps. It contains both a local management router and a collector
router (backbone routers). The collector router works as ABR
III. RELATED WORK holding independent topology database for each area. Second
Tanvi Mehra Proposed algorithm [11] to secure access in type: Is a totally stubby area (area 1) that contains smart
the advanced metering infrastructure communication network meters router (internal routers) that is known as the only route
by using 4-way handshake to authenticate a new smart meter inside area 1 (intra-area) in addition to the default route to go
with authentication server, authenticated smart meter will get outside area 1 through ABR (inter-area). This is to conserve

Third World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4) 173
 bandwidth and reduce CPU loads. In both areas, "set hello" All devices have their own database and consist of a client
interval sends every 10 second between neighbors and dead table: Used to save and retrieve authenticated neighbors (ID,
interval 40 second with cryptographic authentication MD5 public key, AES key, client types (type 1 for collectors, type
hash using specific authentication key that we suppose is set 2 for smart meters)) in AS, (ID and Public key) in collector,
in AS before installation. Adjacency will not have done if (ID, Public key) in the smart meters. Client root table: To save
there are any different parameters between neighbors as route to other authenticated devices connected directly or via
shown in Fig. 4. another authenticated device. My property table: To save and
retrieve owned information (AES key (Kas), private key
(PRas), public key (PUas), DNA Sequence (S), ID (IDas)) in
AS, (AES key (Kc), private key (PRc), public key (PUc), ID
(IDc)) in collector, (AES key (K), private key (PR), public key
(PU), DNA Sequence (S), ID (ID)) in smart meters. Message
Table: To store (smart meter ID, date, consumption reading)
in AS, received message (date, data, type (unicast type 1,
broadcast type 2, multicast type 3)) in smart meters.
1) Authenticate Collector and Smart Meters: This
authentication is accomplished to provide access to the AMI
according to credential founding in local management
authentication server.
a) The Collector authentication: Collector owned Kc,
IDc burned in main chip during manufacturing. AS register
Kc, IDc in owned client table and operators enter client type
(collectors=type1) Before installing the collector. When the
collector starts to run send IDc, and hash (SHA-256) of IDc
and Kc to AS. AS does the same hash calculation. If two
hashes are equal, AS adds IDc in client root table and sends
possess IDas and PUas (RSA 1024) to collector. Collector
Fig. 4. OSPF build and keep-alive neighbours flowchart stores PUas, IDas in client table, and generates owned PRc
and PUc keys then sends PUc with IDc to AS. AS stores
B. Traffic Filtering Level Collector PUc in the client table record.
The objective of traffic filtering is to allow legal traffic b) The smart meter authentication is directly
to pass and drop illegal traffic. Access control list (ACL) is connected to the collector: Same as what happened in hash
applying over routers to filter ingress traffic to prevent DOS collector calculation at start up with the different that client
attack. ACL used to deny Internet Control Message Protocol type is 2. Authenticated collector adds his IDc to help AS
determine the place of smart meter then signs message. AS
(ICMP) packets with no echo-reply from the node's network.
verifies sign If it right, verify smart meter hash then add smart
ACL is composed of a set of permits and denies conditional
meter ID1 in the client root table that route through collector
statements that processed top-down by router operating
IDc. Respond to collector signed message contains of DNA
system Fig. 5 illustrates the ACL packet filtering. Sequence (S) encrypted by K1. Collector verifies the sign and
adds his IDc, PUc to the original message and sends it back
to the requested smart meter. Smart meter decrypts the
message by K1 and store IDc, PUc in client table, S in my
property table. Generate possesses PR1, PU1 and sends ID1,
PU1 to collector. The collector adds ID1, and PU1 in it is
client table and ID1 in client root table.
c) Smart meter authentication which is connected to
collector throws another smart meter: As shown in Fig. 6,
Authenticated smart meter adds the requested meter ID2 in
client root table and adds his ID1 then signs the message.
Collector verifies sign and does the same previews process.
AS verifies sign and requests a meter hash. Respond signed
message contains a DNA Sequence (S) encrypted by K2.
Collector verifies sign and adds his IDc, PUc then signs
Fig. 5. ACL traffic filtering flowchart message back it to the authenticated smart meter.
Authenticated smart meter verifies sign then adds its ID1 and
C. The Proposed End-to-End Security Protocol Level backs it to the requested smart meter. Requested Smart meter
There are three goals for this level. First: Authenticate decrypts message by using K2 and stores IDc, PUc in client
collectors and smart meters, second: Secure real time table and When finds ID1 known it is not directly connected
aggregates data that is transferred from the smart meters to the to collector, S in my property table. Requested Smart meter
autonomous system and last is the secure commands that are responds to collector by its ID2, PU2.
sent from AS to the smart meters.

174 Third World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4)
 Fig. 6. Smart meter authentication throws another smart meter

2) Secure Real Time Aggregate Data: Smart meter c) Broadcast: AS encrypts the message and adds
encrypts its reading through converting consumption reading timestamp then signs message. Collector verifies sign and
and DNA Sequence (S) to binary then inserts first one bit resigns message then sends it to all smart meters connected
from binary of consumption reading then first one bit from directly to it. Smart meter verifies sign and if it is found of
DNA Sequence (S) binary and so on. It converts the binary type 2, it takes copy for itself and resends the message to a
result to a string form then adds the timestamp and the own neighbor connected directly to prevent loop occurrences.
IDx, then signs the message. The collector verifies the sign V. EXPERIMENTAL RESULTS
and adds IDc then resigns the message. AS verifies sign and
process timestamp then reverses bit-wise operation to The proposed model is applied on the topology in Fig. 8,
The experiments time are 120 second. Platform server Dell
decrypts messages and using IDx to adding reading data in
model T7500 with Intel Xeon(R) E5620 2.40 GHZ (2
message table with date. processor) and 32G RAM under windows 10 pro 64-bit
3) Secure Commands: Transition using three methods. operating system. Using eve (emulated virtual environment)
a) Unicast: AS brings destination smart meter IDx and version 2.0.3 to emulate topology with cisco C7200 IOS
encrypts messages through bit-wise operation and add IDc router and Wireshark version 3.0.1 to capture traffic. VMware
finally sign message. The collector verifies sign and resigns version 14 pro with ubuntu 16.04.4 64-bit to virtual
message again to send it to smart meter. The smart meter x autonomous system (service provider) machine this setting is
verifies sign and reverses bit-wise operation to decrypt memory 5G, 2 processor, and 20G HDD, the collector virtual
message. If there is a message from type 1 the meter machine setting is memory 4G, 2 processor, and 20G HDD,
recognizes it and stores the message with its received date. and the smart meters virtual machine settings are memory 3G,
1 processor, and 20G HDD. Proposed E2E security protocol
b) Multicast: As shown in Fig. 7, AS encrypts the
developed using java 8 with spring tool suite release 3.9.2 and
message, adding timestamp and destinations ID1, ID2 and using MySQL version 5.7.21 for database with workbench
collector IDc then signs message. Collector verifies sign and version 6.3.10.
resigns the message again then sends it to the smart meter
ID1. message from type 3, which contains other ID2 and
found in meter client root Table, the meter knows this
message is for itself and another meter connected to it. so, it
sends a copy to other. In case all meters are connected directly
to the collector, the collector sends a copy to every meter as
unicast but the message remains of type 3.

Fig. 7. Transmission of multicast command

Fig. 8. Implementation toplogy with attack scenario

Third World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4) 175
 A. Experiment 1: Examine hello interval that packet size is C. Experiment 3: Performance consumption reading and
1072 bits was authenticated using MD5 between AS, management commands
collector, SM1, SM2 and SM3 routers In the proposed model founding the higher throughput
Using point to point connection between AS and collector, occurs in the second 40 of the experiment time. It was 2532
collector and SM1, SM1 and SM2, collector and SM3. We bytes bidirectional transferred between utility and smart
notice that interval does not exceed dead time and topology meters 2 and 3 to process the meters consumption reading. The
routers are in adjacency state as shown in Fig. 9. lowest throughput occurs in the second 47. It was 1133 bytes
bidirectional transferred between utility and smart meter 1 to
process unicast command message as shown in Fig. 12.

Fig. 9. Hello interval

B. Experiment 2: With/without using ACL to drop ICMP

packets
Fig. 12. Amount of Information Send and Received in the proposed model
Suppose that an attacker tries to down SM2 gateway (DOS
attack) as shown in fig. 8, by using ICMP packets. Without The most one delay has reached to 137.9 millisecond
applying any traffic filtering over SM2 gateway, notice SM2 occurs in smart meter 1 that waiting for multicast respond
gateway is fill up of requests and engaging in response to from smart meter 2 in the second 66 of the experiment time.
attacker that exploit SM2 gateway CPU, memory and The least delay has reached to 35.98 millisecond occurs in
bandwidth as shown in Fig. 10.
smart meter 1 that waiting for success response for it is
consumption as shown in Fig .13.

Fig. 10. ICMP DOS attack Fig. 13. Delay chart using The proposed model

After applying ACL over SM2 gateway with no echo- D. Experiment 4: Defend AMI Using the Proposed Model to
reply, notice all requests are dropping as shown in Fig .11. Counter Attacks
a) Replay attack: Proposed model faces this type of
attack by using timestamp. When the smart meter prepares its
message adding timestamp then signs the message and sends
it to the provider. After the service provider receive and
verifies message subtract its time from message time. If the
result is less than delta time so the message is normal message
other than the message is replay attack message and the
message will drop.
b) Modifying reading: Proposed model faces this type
of attack by using hash function. When the smart meter
prepares its message adding timestamp then sign message and
sending it to provider the collector receive message and check
the sign by decrypt using meter public key and recalculate
hash again. If recalculated hash dose not equal the message
Fig. 11. Filtering ingress ICMP packets

176 Third World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4)
hash so, the message is dropped because it is altered during TABLE III. PROPOSED MODEL THROUGHPUT, DELAY TIME AND
ENCRYPTION/DECRYPTION TIME COMPARED WITH AES ALGORITHM
transmission.
Proposed Model (DNA) Using AES Algorithm
c) Eavesdropping attack: The attacker captures the
message to know the information about the customer but

Throughput

Throughput
proposed model can face this type of attack by encrypt

Encryption

Encryption
Decryption

Decryption
Max Delay

Max Delay
Time (SP)

Time (SP)
Nodes

Time

Time
Max

Max
reading using DNA bit-wise operation to keep user privacy
and achieve confidentiality.
E. The Security Analysis
Bytes MS μ μ Bytes MS μ μ
Present set of attacks that the proposed model can detect
and deny. These types of attacks and authentication compared
SP
with the previous work as shown in Table 1.

TABLE I. COMPARISON BETWEEN OTHERS AND PROPOSED MODEL

227.03

327.59
88.75

458.6
SM1
Authentication Attacks

137.9
2532

3751

1076
Eavesdropping
Modification

445.92

408.64
Scheme

84.67
SM2
gateway
Mutual

303
Replay

DOS

232.72

478.81
91.85

229.4
SM3
[11] √

[32] √ √ √

[31] √ √ √ √ VI. CONCLUSION AND FUTURE WORK

[33] √ √ √ This article proposes a new paradigm of electricity grid
using information and communication technology (ICT) for
[34] √ √ √ √ √
a smart and automated design. In this article, combat threats
[35] √ √ √ √ √ that confront advanced metering infrastructure were
Proposed
√ √ √ √ √ √
discussed. A proposed model that authenticates routers and
Model uses ACL to prevent unneeded traffic from flow inside the
network is explained. The mutual authentication between the
The security analysis of the model shows that the proposed local distribution management and the remote devices such
model provides a more secure and efficient transmission of as collector and smart meters are also highlighted. The
data collection, a better management of messages and better cryptography overhead is lightweight. The security analysis
authentication computation cost. effective in bidirectional of the model shows that this proposed model provides a more
data transfer (throughput and delay) because it does not secure and efficient transmission of data collection, and a
consume too much bandwidth compared to AES algorithm better management of messages. Future work will enhance
and the cryptography performance for consumption reading delay time, compare performance (throughput, delay) with
messages is better than the AES algorithm as shown in Table different communication technologies, and Implement
2 and Table 3. proposed model over a real hardware environment (testbed)
and compare them with the current results.
TABLE II. MUTUAL AUTHENTICATION COMPUTATION COST WITH
OTHERS
ACKNOWLEDGMENT
Scheme SM2 SM1 Collector SP Total Special thanks to everyone who contributed to the
Cost
7SED + 7SED + 14SED +
implementation and testing of this work and to northern zone
[31] 7H + 7H + 14H + in the Egyptian Electricity Transmission Company (EETC).
4ASED 4ASED 8ASED
REFERENCES
1SED + 1SED + 4SED +
[32] 2SED
1MAC 1MAC 2MAC
3T + 4H 3T + 5H 6T + 9H [1] Fang X, Misra S, Xue G, Yang D, “Smart Grid - The New and
[35] + 2MAC + 2MAC + 4MAC Improved Power Grid: A Survey,” IEEE Communications Surveys &
+ 2SED + 2SED + 4SED Tutorials, vol. 14, pp. 944-980, Dec. 2011.
1H + 2H + 4H + [2] Delgado-Gomes V, Martins JF, Lima C, Borza PN, “Smart grid
Proposed 1H + 1SED + 3SG + 2SED + 4SED + security issues,” in IEEE International Conference on Compatibility
Model 1SED 1SG + 3SV 2SG + 6SG + and Power Electronics (CPE), 2015, pp. 534-538.
1SV 2SV 6SV [3] Kabalci Y, “A survey on smart metering and smart grid
T – Point Multiplication; H – one-way Hash; SED – Symmetric encryption and decryption; ASED communication,” Renewable and Sustainable Energy Reviews, vol.
Asymmetric encryption and decryption; SG – Signature Sign; SV – Signature Verification; MAC – 57, pp. 302-318, May. 2016.
Message Authentication Code
[4] Enos N, “Implementing an Integrated Security Management
framework to ensure a secure Smart Grid,” in IEEE International
Conference on Advances in Computing, Communications and
Informatics, 2014, pp. 778-784.
[5] Sophia Kaplantzis S, S_ekercio_glu YA, “Security and smart
metering,” in IEEE European Wireless Conference, 2012, pp. 1-8.

Third World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4) 177
[6] Komninos N, Philippou E, Pitsillides A, “Survey in Smart Grid and [20] Rawat DB, Bajracharya C, “Cyber Security for Smart Grid Systems:
Smart Home Security: Issues, Challenges and Countermeasures,” Status, Challenges and Perspectives,” in IEEE SoutheastCon, 2015,
IEEE Communications Surveys & Tutorials, vol. 16, pp. 1933-1954, pp. 1-6.
April. 2014. [21] Stallings W, Cryptography and Network Security: Principles and
[7] Wenye W, Lu Z, “Cyber security in the smart grid: survey and Practices. Upper Saddle River, NJ: Pearson, 2006.
challenges,” Computer Networks, vol. 57, pp. 1344-1371, April. 2013. [22] Santos O, Stuppi J, CCNA security 210-260 official cert guide,
[8] Ur-Rehman O, Zivic N, Ruland C, “Security Issues in Smart Metering Indianapolis: Cisco Press, 2015.
Systems,” in IEEE International Conference on Smart Energy Grid [23] Rodriguez FJ, Fernandez S, Sanz I, Moranche Ml, Bueno EJ,
Engineering (SEGE), 2015, pp. 1-7. “Distributed Approach for SmartGrids Reconfiguration based on the
[9] Ur-Rehman O, Natasa Zivic N, “Secure Design Patterns for Security OSPF routing protocol,” in IEEE Transactions on Industrial
in Smart Metering Systems,” in IEEE European Modelling Informatics, 2016, pp. 864-871.
Symposium, 2015, pp. 278-283. [24] Odom W, CCNA Routing and Switching: ICND2 200-105 Official
[10] Rodriguez-Diaz E, Palacios-Garc__a EJ, Savaghebi M, Vasquez JC, Cert Guide, Indiana: Cisco press, 2017.
Guerrero JM, Moreno-Munoz A, “Advanced Smart Metering [25] Wallace K, CCNP Routing and Switching: ROUTE 300-101 Official
Infrastructure for Future Smart Homes,” in IEEE International Cert Guide, Indiana: Cisco press, 2015.
Conference on Consumer Electronics (ICCE-Berlin), 2015, pp. 29-31. [26] Ming-hao W, “The Security Analysis and Attacks Detection of OSPF
[11] Mehra T, Dehalwar V, Kolhe M, “Data Communication Security of Routing Protocol,” in IEEE International Conference on intelligent
Advanced Metering Infrastructure in Smart Grid,” in IEEE 5th computation technology and automation (ICICTA), 2014, pp. 836-839.
International Conference and Computational Intelligence and [27] (2000) the IETF website. [Online]. Available:
Communication Networks, 2013, pp. 394-399. https://www.ietf.org/rfc/rfc2827.txt
[12] Anzalchi A, Sarwat A, “A Survey on Security Assessment of [28] (1999) the IETF website. [Online]. Available:
Metering Infrastructure in Smart Grid Systems,” in IEEE https://www.ietf.org/rfc/rfc2644.txt
SoutheastCon, 2015, pp. 9-12. [29] Majumder A, Majumdar A, Podder T, Kar N, Sharmas M, “Secure
[13] Skopik F, Ma Z, “Attack Vectors to Metering Data in Smart Grids Data Communication and Cryptography Based on DNA Based
under Security Constraints,” in IEEE International Conference on Message Encoding,” in IEEE International Conf on Advanced
Computer Software and Applications Workshops, 2012, pp. 134-139. Communication Control and Computing Technologies, 2014, pp. 360-
[14] Jafary P, Repo S, Koivisto H, “Secure Integration of the Home Energy 363.
Management System to the Battery Management System in the [30] Roy B, Rakshit G, Singha P, Majumder A, Datta D, “An improved
Customer Domain of the Smart Grid,” in IEEE PES General Meeting Symmetric key cryptography with DNA based strong cipher,” in IEEE
| Conference & Exposition, 2014, pp. 1-5. International Conf on Devices and Communications, 2011, pp. 1-5.
[15] Blilat A, Bouayad A, Chaoui N, El Ghazi M, “Wireless Sensor [31] Kumar V, Hussain M, “Secure Communication for Advanced
Network: Security challenges,” in IEEE Wireless Sensor Network: Metering Infrastructure in Smart Grid,” in IEEE India Conference,
Security challenges, 2012, pp. 68-72. 2014. pp. 1-6.
[16] Jafary P, Repo S, Koivisto H, “Secure Communication of Smart [32] Yan Y, Hu RQ, Das SK, Sharif H, Qian Y, “An Efficient Security
Metering Data in the Smart Grid Secondary Substation,” in IEEE Protocol for Advanced Metering Infrastructure in Smart Grid,” in
Innovative Smart Grid Technologies - Asia (ISGT ASIA), 2015, pp. 1- IEEE Network, 2013, pp. 64-71.
6. [33] Ozgur U, Tonyali S, Akkaya K, “Testbed and Simulation-Based
[17] Lu Z, Lu X, Wang W, Wang C, “Review and Evaluation of Security Evaluation of Privacy-Preserving Algorithms for Smart Grid AMI
Threats on the Communication Networks in the Smart Grid,” in IEEE Networks,” in 41st IEEE Conference on Local Computer Networks
Communications Conference – MILCOM, 2010, pp. 1830-1835. Workshops (LCN Workshops), 2016, pp. 181-186.
[18] Modares H, Salleh R, Moravejosharieh A, “Overview of Security [34] Tonyali S, Akkaya K, Saputro N, Cheng X, “An Attribute & Network
Issues in Wireless Sensor Networks,” in IEEE Third International Coding-Based Secure Multicast Protocol for Firmware Updates in
Conference on Computational Intelligence, Modelling & Simulation, Smart Grid AMI Networks,” in 26th IEEE Conference on Computer
2011, pp. 308-311. Communication and Networks (ICCCN), 2017, pp. 1-9.
[19] Wang Y, Lin W, Zhang T, “Study on Security of Wireless Sensor [35] Kumar P, Gurtov A, Sain M, Martin A, Phuong, Ha, “Lightweight
Networks in Smart Grid,” in IEEE International Conference on Power Authentication and Key Agreement for Smart Metering in Smart
System Technology, 2010, pp. 1-7. Energy Networks,” IEEE Transactions on Smart Grid (early access),
2018.

178 Third World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4)