SEPS-AKA: A SECURE EVOLVED PACKET SYSTEM AUTHENTICATION AND KEY AGREEMENT SCHEME FOR LTE-A NETWORKS

Natarajan Meghanathan et al. (Eds) : WiMONe, NCS, SPM, CSEIT - 2014
pp. 57–70, 2014. © CS & IT-CSCP 2014 DOI : 10.5121/csit.2014.41206
SEPS-AKA: A SECURE EVOLVED PACKET
SYSTEM AUTHENTICATION AND KEY
AGREEMENT SCHEME FOR LTE-A
NETWORKS
Zaher Jabr Haddad1
, Sanaa Taha2
and Imane Aly Saroit Ismail2
1
Department of Computer Science, Faculty of Applied Science,
Al-Aqsa University, Gaza, Palestine
zj.haddad@alaqsa.edu.ps
2
Information Technology Department, Faculty of Computers and Information,
Cairo University, Cairo, Egypt
staha@fci-cu.edu.eg, i.saroit@ fci-cu.edu.eg
ABSTRACT
In this paper, we propose a secure authentication of the Evolved Packet System Authentication
and Key Agreement (EPS-AKA) for the LTE-A network. Our scheme is proposed to solve the
problem of sending the IMSI as a clear text, and hence prevents the mobility management entity
attack. We will use public key (PK) cryptography to protect the transmitted messages, the RSA
scheme computation to compute a temporary value to the IMSI, and nonce to generate
challenge messages toward the opposite side. Our scheme does not need to change the original
framework and the infrastructure of the LTE-A network, although a ciphered IMSI is
transmitted. The authentication procedure is performed by the HSS to authenticate the UEs and
the MME; therefore, the impersonating of the MME and UEs is not possible. Our evaluation
demonstrates that the proposed scheme is secure and achieves the security requirements of the
LTE-A subscribers such as privacy, authentication, confidentiality and integrity. In our scheme,
we try to maintain the problems defined in the previous related works.
KEYWORDS
Long Term Evolution – Advanced, Authentication and Key Agreement, Home Subscriber Server,
Mobility Management Entity, User Equipment.
1. INTRODUCTION
The Long Term Evolution-Advanced (LTE-A) network is a packet based system specified by the
Third Generation Partnership Project (3GPP) towards fourth-generation (4G) mobile; in order to
meet more subscriber needs. Among those communications, LTE-A is the next generation of the
cellular communication system that meets more subscriber needs, such as: 1) wider bandwidth
that supports up to 100MHz via aggregation of 20 MHz blocks, 2)Multi Input Multi output
(MIMO) that allows the use of multiple antennas in the transmitter and the receiver in order to

58 Computer Science & Information Technology (CS & IT)
improve the communication performance, 3) coordinate Multiple transmission (CoMP) that
allows coordinates scheduling, beam-forming and joint processing transmission, 4)Heterogonous
Network (Het-Net) that supports enhanced inter-cell interference coordinate (eICIC) to deal at the
interference issues at the cell edge, and 5)relaying capabilities that achieves self-backhauling of
the radio signal between a base station and User Equipment (UE) [1].
The EPS-AKA is the authentication protocol used by the LTE-A network to perform the
authentication and key agreement security services. EPS-AKA protocol was improved to prevent
malicious attacks such a redirection, rogue base station, and Man in the middle attacks. A
malicious attack can be any action intents of acquiring, destroying, modifying or accessing a
transmitted data without permission. However, the lack of privacy and denial of services attack
still a big weakness of the EPS-AKA protocol. This LTE-A's security weakness is represented in
the processes of registration, synchronization failure, and roaming to a new mobility management
entity (MME), when the MME requests the international mobile subscriber identity (IMSI) of the
User Equipment (UE). Therefore the IMSI disclosure may incur severe problems [2].
Many attacks may violate the vulnerabilities of the authentication in the LTE-A network such as
[3][4]:
a) Replay Attack, which attains to perform maliciously or fraudulently repeated or delayed
transmitted messages in order to increase the flow in the network and therefore, may make
system toppled [3][4].
b) Denial of service (DoS) attack, which attains to make a machine or network resources
unavailable to legitimate users [3][4].
c) Man in the Middle (MITM) attack, which makes independent connection between two
victims in order to intercept or inject fake messages [3][4].
d) Impersonation attack, which attempts to use a fake identity to gain unauthorized access to
network system through legitimate access identification [3][4].
In this paper, a novel scheme is proposed to solve the problem of sending the IMSI as a clear text,
and hence prevents the mobility management entity attack. In our scheme, we will use three
levels of security. First, nonce is used to generate challenge messages toward the opposite side.
Second, PK cryptography is used to protect the transmitted messages. Third, the RSA scheme
computation is used to compute a temporary value to the IMSI.
The remainder of this paper is organized as follows: Section II illustrates the related work. The
system models, including network, threat, and trust models, are presented in Section 3. In section
4 the preliminaries are discussed. In section 5, the proposed system, SEPS-AKA, is introduced. In
sections 5 and 7, the security analysis and performance evaluation are discussed, respectively. In
section 8, conclusion and future work are provided.
2. RELATED WORK
In [5], Park et al., introduce number of possible security risks may be caused due to the open
nature of the 4G networks. First, a large number of external connectivity points with peer
operator, third-party applications providers, the public Internet, and with numerous heterogeneous
technologies accessing the infrastructure, serves as potential security holes if the security
technologies do not fully interoperability. Second, multiple service providers share the core
network infrastructure, meaning that compromise of a single provider may result in collapse of
the entire network infrastructure. Third, service theft and billing fraud can take place if there are
third-parties masquerading as legitimate ones [5]. New end-user equipment's can also become a
source of malicious (e.g., DoS) attacks, viruses, worms, spam mails and calls. In particular, the

Computer Science & Information Technology (CS & IT) 59
Spam over Internet Telephony (SPIT), the new spam for VoIP [5], becomes a serious problem
just like the e-mail spam today. For example, SPITs targeting VoIP gateways can consume
available bandwidth, thereby severely degrading QoS and voice quality. Clearly, the open nature
of VoIP makes it easy for the attackers to broadcast SPITs similar to the case of spam emails.
Other possible VoIP threats include: (1) spoofing that misdirects communications, modifies data,
or even transfers cash from a stolen credit card number, (2) SIP registration hijacking that
substitutes the IP address of packet header with attacker's own, (3) eavesdropping of private
conversation that intercepts and crypt-analyzes IP packets, and (4) phishing attacks that steal user
names, passwords, bank accounts, credit cards, and even social security numbers.
In [6], Purkhiabani et al. propose a new scheme to preserve the privacy of the IMSI by encrypting
it using a temporary random number (MSR) during its transmission into the EUTRAN interface.
In addition, HSS generates only one Authentication Vector (AV) to use in each authentication
process in order to preserve the bandwidth of the CN. Therefore, the UE sends a message, which
contains a concatenation of MSR, IMSI, and MSMAC, to the MME, which in turn forwards this
message to the HSS. The MSR is a random number generated by the UE, and MSMAC=
f1k(MSR), where f1 is a cryptographic function used to generate 128-bit output using 128-bit
input key. After receiving the transmitted message, the HSS verifies the IMSI, generates and
sends back one AV to the MME, and hence, the original authentication and key agreement are
performed. In this scheme, the use of same framework of the original LTE-A authentication
scheme, decreases the HSS bandwidth consumption, the protection of the IMSI, and bandwidth
preserving of the CN. But as the MSR is generated by the UE, this increases the possibility of
malicious UE and MME. Moreover, the bandwidth consumption is moved from the CN to the
radio interface.
In [7], Hamandi et al. propose a scheme to solve the privacy problem in the LTE-A authentication
scheme to prevent the masquerading of the MME. Authors employ the public key infrastructure
to provide more powerful MME and HSS elements. MME generates and sends a random number,
RANDMME, to the UE to compute a vector of parameters, and then returns a message to the
MME, which in turn adds its identity and digital signature and forwards the message to the HSS.
At receiving, the HSS verifies the identities of both the MME and IMSI, and then generates a new
random mobile subscriber identity, RMSI, to concatenate with the AVs. Both AVs and RMSI are
sent back to the MME in order to complete the original authentication and key agreement
procedures. In this scheme, a ciphered IMSI is sent, a virtual number (TMSI) is used in the next
hops instead of the IMSI, and legal MME identity is protected by digital signature. However,
EUTRAN consumption is increased. In addition, the initiation procedure is started from the
MME, which also allows the possibility of the presence of malicious MME. Additionally, this
scheme is not integrated with the original mobility procedures, such as handover and localization
[7].
In [8], Abdo et al. define four security weaknesses in the original LTE AKA protocol: IMSI
catching, tracking user temporary identity due to linkability and security network authentication.
In addition, the authors propose two countermeasures to use in order to solve these problems:
Public Key Infrastructure (PKI) and pseudonyms based approaches. The advantage of this work is
the security capabilities that are performed using the PKI. However, there is a critical problem
that is the first hop dependency, where the UE depends on a pre-stored cipher Key (CK) and
identity Key (IK) to generate the initial pseudonyms. CK and IK are generated by the pre-shared
cryptographic function using the pre-shared secret key (K) between UE and HSS and a random
challenge RAND that is generated by the HSS, therefore, the HSS should performs some
computations before initialization phase, and surely this depends on the IMSI.

In [9], Abdo et al. propose a scheme called EPS mutual authentication and Crypt-analyzing (SP-
AKA), which is a self-certified based protocol, in order to solve the positive capturing of the
IMSI during user identification and key agreement protocols. Authors use the PKI to encrypt the
transmitted AKA messages. Hence, provide a high security level, but the fake MME is still a
problem.
In [10], Lai et al. propose a new scheme for group base communication authentication called a
secure and efficient group authentication and key agreement protocol for LTE networks (SE-
AKA). This scheme uses the Elliptic curve Diffie-Hellman to achieve the key forward/backward
secrecy and also adapts asymmetric cryptosystem to protect user privacy. For group
authentication, SA-AKA uses a group temporary key (GTK), which employs a well-known keys
generation algorithm, Diffie-Hellman, and also provides a strong security level where subscriber
must meet the restrictions of the authenticated group, before network authentication. The
problems of this scheme are the consumption of the MME where the Elliptic curve Diffie-
Hellman consumes time to generate and distribute the public keys between group members, while
the main role of the MME is to work as a gateway between the HSS and the UEs. Also, the
proposed group is considered as an uncontrolled area in the network and used to break the
security of the network since the authentication permissions are invoked to the group authority
instead of the HSS.
In [11], Zheng el at. propose a hybrid AKA scheme that uses a trusted model platform and PKC
to adapt the AKA. This scheme uses a password associated with fingerprint and PKC to achieve
the authentication between the UE and HSS.
In our scheme, we try to maintain the problems defined in the previous related works. Our
proposed scheme does not need to change the original framework and the infrastructure of the
LTE-A network, although a ciphered IMSI is transmitted. The authentication procedure is
performed by the HSS to authenticate the UEs and the MME; therefore, there is no possibility for
any occurrence of fake MME and UEs.
3. SYSTEM MODEL
In this section, we describe the system models for the SEPS-AKA including, the network model
and the threat model.
3.1. Network Model
In this subsection, we will explain the LTE-A network architecture and the original authentication
and key agreement protocol used in the LTE-A network
3.1.1 LTE-A Network Architecture
The architecture of the LTE-A network is mainly composed of two components as depicted in
Figure 1; the Evolved Packet Core (EPC) and Evolved Universal Terrestrial Radio Access
Network (EUTRAN). [2][16].
The EPC represents the wired part in the network, which is responsible for the overall control of
UEs and the bearer establishment. Each entity in the EPC has a responsibility as follows;
• MME to manage bearer and connection.
• HSS to maintain the user subscription data and MME identities.
• Packet Data Network (PDN) and Packet Data Network Gateway (PGW) to perform mobility
anchor and internetworking within the 3GPP and non-3GPP technologies respectively.

• Policy Control and Charging Rule function (PCRF) to control decision making of flow and
Quality of Service (QoS).
EUTRAN is the Radio Access Network (RAN) in the LTE-A system; mainly two components are
included [2][16];
• UE which is the mobile phone handset.
• Evolved Node B (eNB) presents the Base Transceiver System (BTS) and the Base Station
Subsystem (BSS) in the non-3GPP technologies.
Figure 1: LTE-A Network Model
Number of eNBs is interconnected via the EUTRAN to manage multiple cells and to allow the
interaction between different protocol layers in order to perform the radio resource management,
header compression, securely capabilities and connectivity functions.
The security of cellular communication is a very important issue to subscribers. Attacker can
exploit any security flaw to perform their goals. Telecommunication, video and audio streaming,
mobile banking, data transmission, commences and etc. are a mobile application that may be
attacked and hence causes a sophisticated problems. Authentication is the security issue that
verifies users to the network. The problem in this issue allows unauthorized users to access the
network.
Consider the system and the communication models of the LTE-A as depicted in Figure 1, the
MMEs are the connection link between the HSS and the UEs. In our propose scheme, we do not
introduce any modification on the original infrastructure of the LTE-A. Only we introduce a soft
modification on the parameters and algorithms that are adapted in the system entities; such as the
RSA scheme in the UEs and the HSS, to generate and regenerate parameters, and the certificate
authority, to distribute PKs to between MMEs and UEs [2][16].
3.1.2. LTE-A Authentication Procedure
EPS-AKA authentication procedure was proposed in the 3GPP release 9 for LTE networks. EPS-
AKA can broadly be divided into two stages: (1) authentication data distribution, and (2) user
authentication and key agreement. The former enables the home network (HN) of a mobile
equipment (ME) to distribute authentication data to the serving network (SN) where the ME
device is visiting. The latter is to establish new session keys between the ME and the SN. The
EPS-AKA protocol works as follows [10]:

a) An UE sends an access request message to the MME.
b) Upon receiving a request, the MME launches an authentication procedure by asking the UEs
identity (IMSI).
c) In response to the MME, the UE sends its identity (IMSI).
d) The MME sends an authentication data request message containing IMSI to the HSS for
acquiring Authentication Vectors (AVs).
e) The HSS first generates AVs for the MME, an AV comprising a RAND, XRES, AUTN and
KASME in steading of IK and CK in UMTS AV. The AV is expressed as AV =
RAND||XRES||KASME||AUTN, and AUTN = SQN ⨁(AK || AMF || MAC), where ⨁ is a
simple bitwise XOR and || is a simple concatenation operations.
f) The HSS sends back an authentication data request message including the generated AV (for
the corresponding UE), so that the MME is authorized to authenticate the requesting UE.
g) Upon receipt of authentication vectors, the MME sends RAND and AUTN piggy-backed on
authentication request to the UE, enabling the ME to verify the correctness of SQN and
compute the RES.
h) The UE verifies the correctness of SQN by computing MAC and comparing it with the MAC
carried in AUTN. If matched, the ME computes and sends the corresponding response RES
back to the MME in an authentication response message.
i) Once the MME receives and verifies RES correctly, it chooses the corresponding KASME as
the session key to protect its communication with the ME. In addition, the ME calculates its
KASME accordingly.
3.2. Threat and Trust Model
The violation of the wireless network systems is a common target of hackers, thus, in this
subsection; we consider two type of attacks that may violate the security of the LTE-A system
such as the cyber-attack and the side channel attack [12].
Cyber-attack is any type of offensive maneuver employed by hackers that targets computer
information systems, infrastructures, and computer networks by various means of malicious
actions usually originating from an anonymous source that steals, alters, or destroys a specified
target by hacking into a susceptible system. Cyber-attacks can range from installing spyware on
computer systems to attempts to destroy the infrastructure of entire nations. Cyber-attacks have
become increasingly sophisticated. In the LTE-A system a cyber-attack may be employed as a
malicious such as, MME, UE, Home eNB (HeNB), and non-3GPP access point, in order to break
down the system [12].
A side channel attack may be violate the security of the LTE-A system, since it relies on the
relationship between information emitted through the side-channel and the secret data depending
on information gained from the physical implementation of a cryptosystem. In the LTE-A
network, femto and micro cells are good environments for side channel attack to be efficient
where a technical knowledge of the internal operation of the system and powerful statistical
methods are defined [12].
The trusted model is a TTP in the PKI that creates the public/private keys [12]. However, a
meaningful trust model for a PKI must consider the semantic assumption and human cognition of
trust relationship, such as the legal constricted agreements between participants and how identity
information is displayed and represented. In our proposed scheme, we use the Pretty Good
Privacy (PGP) to authenticate UE to the HSS. PGP [12] is a free version commercial encryption
entity used to authenticate parities, the PGP allows every user to play a role of relaying parity
where each one can sent certificate to each other. Therefore, PGP defines three methods for users

and relaying party to obtain a public key of other users; 1) a secure out-of-band channel, such as
physical meeting, 2) online trust decision based on introductions of new certificate from
previously trusted users, and 3) based on discretionary trust decision when receiving a public key
[12].
4. PRELIMINARIES
In this section, both the public key infrastructure and the RSA scheme are presented as our
preliminaries, since the SEPS-AKA is based on both of them.
4.1. Public Key (PK) Cryptography
It is asymmetric cryptography involving the use of two separate keys, unlike symmetric
encryption, that uses only one key [13][14]. The use of two keys has profound consequences in
the areas of confidentiality, key distribution, and authentication. Public-Key Cryptography was
developed to address two key issues. First, key distribution, in how to have secure
communications in general without having to trust a KDC with your key. Second, digital
signatures, in how to verify a message comes intact from the claimed sender. Public-key
cryptography involves the use of two keys. First, Public-key is known to everybody in order to
use for encrypting messages, and verifying signatures. Second, Private-key, known only to the
recipient, used to decrypt messages, and sign (create) signatures. Public key cryptography
applications are classified into three categories: 1) Encryption/decryption; the sender encrypts a
message with the recipient's public key, 2) Digital signature; the sender signs a message with its
private key, either to the whole message or to a small block of data that is a function of the
message, and 3) Key exchange, two sides cooperate to exchange a session key. The main
advantage of public key cryptography is the asymmetricity since who encrypts message or
verifies signature cannot decrypt same messages or create same signatures, therefore, it is
infeasible to determine private key from public [13][14].
4.2. The RSA Scheme
The security principle of the RSA scheme is based on hardness of the factorization problem due
to the cost of factorizes large numbers [13][14]. Each user generates a public/private key pair by
selecting two large prime numbers at random: p, q. compute n = p * q and ø(n)=(p-1)(q-1).
Randomly, the RSA scheme selects an odd number e, where 1< e < ø(n), gcd(e , ø(n)) =1, e*d=1
mod ø(n), and 0≤ d ≤n. Then publishes their public key PU={e,n} and keeps secret private key
PR={d,n}. To encrypt a message, M, the sender obtains a public key of the recipient and
computes the cipher text, C = Me
mod n, where 0 ≤ M < n. To decrypt the cipher text C, the
owner of the message uses their private key, PR={d,n}, and computes M = Cd
mod n [13][14].
5. PROPOSED SCHEME (SEPS-AKA)
Figure 2 describes the SEPS-AKA proposed scheme, the workflow of the SEPS-AKA is the
similar to the framework of the original EPS-AKA scheme. The methodology of our proposed
system uses the two methods explained in the previous section to enhance and adapt the privacy
of the original LTE-A authentication procedure. First, the infrastructure of the public key
cryptography is used to encrypt the exchanged data between LTE-A network entities. Second, the
RSA scheme computation is used to compute the used parameters in the previous section. A pre-
shared secret key K is used as the original LTE-A network where the key was industrially preset
to the devices and stored physically in the USIM and to the HSS. The workflow of the SEPS-
AKA scheme is described as follows:

Figure 2: The SEPS-AKA Scheme
5.1. UE MME
In this stage, UE initiates five parameters; p, q, n, d, and L. these parameters are computed based
on the RSA scheme computations using the IMSI and k as follows:
a) Parameter Initiation
o Select large prime number p.
p = IMSI + L (1)
Where L is an integer number computed when the IMSI is not prime as L =
0, if IMSI is prime, otherwise L is the step value of the next prime after
IMSI.
o Set q as a random large prime number
o Compute n :
n = p * q
(2)
o Compute large number d:
d = e-1
mod (p-1) (q-1)
(3)
Where e is an odd number < ( (p-1) *(q-1))
b) Parameter Encryption: Encrypts d using a standard encryption algorithm (AES) to provide
an encrypted parameter, s.
s = Ek (d)
(4)

c) Message Building and originating: Concatenates the last parameters p, n, q, s, L and a
nonce1 together, encrypts the result using the public key of the HSS (PKHSS) and originating
the encrypted message to the HSS.
msg1 = EPKHSS
(e || n || s || q || L || nonce1) (5)
5.2. MME HSS
Once MME receives msg1, it builds msg2 as illustrated in (6), which is the received message
from the UE msg1 plus its identity MMEid and originates the message to the HSS
msg2 = msg1 || MMEid (6)
Where MMEid is the identity of the MME
5.3. HSS MME
Once the HSS receives msg2 from the MME, it decrypts the message and executes the following:
o IMSI extraction
IMSI = (n / q) (7)
• Check the primarility of the extracted IMSI,
If IMSI is prime then
p = IMSI (8)
Otherwise
p = IMSI – L (9)
Where L is an integer number transmitted from the UE, and was computed
when the IMSI is not prime in equation (1) as; L = 0, if IMSI is prime,
otherwise L is the step value of the next prime after IMSI
o IMSI Verification:
Compute d’ as illustrated in (10); a large number computed based on the RSA scheme
computation, in order to verify the IMSI in the HSS side. The IMSI verification in the HSS is
done as follows; check d’ = d , if true then the IMSI is true; this process is not supported in the
original EPS-AKA scheme, while the IMSI is verified in the HSS using database query.
d’ = (e-1
) mod (p-1)(q-1) (10)
After the IMSI verification is done correctly the remained steps are doing normally as the original
EPS-AKA scheme.
6. SECURITY AND PRIVACY ANALYSIS
In this section, we analyze the security of our scheme to demonstrate that it meets the security
requirements of the LTE-A systems. In our scheme, three levels of security are used: PKI, the
RSA scheme computations and nonces. Nonces are random numbers generated by UE, MME and
HSS to use in generating challenge messages toward the opposite side. A different Nonces are
used in each authentication procedure, therefore, the reusing of these Nonces are not efficient. An
out-of-sync situation will lead to authentication failure.
Consider a cyber-attack, in which a malicious UE aims to register to the LTE-A network, the
malicious UE need to gain the computation of the RSA scheme parameters (p, q, s, and n) and to

gain the encryption information, which is required to prepare the message before sending from
the UE. For the malicious MME, at the first registration time, the MME is considered as a
gateway to route encrypted messages from the UE to HSS, while the MME must concatenate its
certificate with the routed message in order to prove its authenticity to the HSS. Therefore, we
consider the presence of the cyber-attack is impossible.
Consider a legal UE is worked through a femto and micro cells, which are two authorized
environments uncontrolled by the LTE-A network, IMSI is not sent through the authentication
message but is computed by the RSA scheme parameters (p and q), which is an NP problem,
while the side channel attacks need a technical knowledge of the internal operation of the system
and powerful statistical methods to be efficient.
In addition, our scheme prevent replay, impersonate attacks, the Man in the Middle, and DoS
attacks. The replay attack is prevented by using the nonces in the transmitted messages, therefore,
it is no possibility to use this message again. In addition of using the PK cryptography to encrypts
the transmitted messages, the IMSI, the legal identity of the UE, is not transmitted in clear text
over the transmitted messages, therefore, the attack cannot able to impersonate the identity of the
UE. The MitM and DoS attacks are prevented as; if a member is able to sniff PKI, it still cannot
computes the IMSI using the RSA scheme computations, although these messages are sent with
PKI protection therefore, the attacker may not be able to hack this data since PKI is having the
residency of DoS and MitM attacks. Therefore, the SEPS-AKA scheme attained the security
requirements such as privacy, confidentiality, authentication, and data integrity.
Table 1. EPS-AKA Security Requirements
Entity mutual
authentication
Privacy Confidentiality Data Integrity
SEPS-AKA Yes Yes Yes Yes
EC AKA Yes Yes Yes No
SP AKA Yes No Yes No
HSK AKA Yes No No No
EPS-AKA Yes No No No
Table 1 is a comparison between SEPS-AKA scheme and the other previous EPS-AKA schemes
in addition to the original EPS-AKA. The SEPS-AKA scheme adopts the same secured
architecture as the EPS-AKA protocol. Therefore, it has the same security threshold in most
situations. As illustrated in Table 1, the SEPS-AKA scheme can attain the security requirements
as follows:
6.1. Entity mutual authentication
All schemes attain the entity mutual authentication since a UE is identified to the HSS by its
IMSI. However, comparing with the SEPS-AKA scheme, where an UE is identified
mathematically by its IMSI, as mentioned in equations (1), (7), (8), and (9), and the mathematical
computation were based on the RSA scheme. The original and the EC-AKA schemes, the
transmission of the IMSI gains the probability to different previous attack. In the SP-AKA and
the HSK-AKA, the first step is begin from the MME that maintain a high probability of the
presence of cyber-attack and the side channel attack.

6.2 Privacy
To ensure user privacy, the IMSI should be confidentiality protected. It should never be
transmitted without protection. The EC-AKA achieves the privacy since an encrypted IMSI is
transmitted. But the remainder schemes has no privacy since the IMSI is transmitted in clear text.
But the SEPS-AKA scheme attains a high level of privacy where the IMSI is protected using the
public key of the HSS.
6.3 Confidentiality
Confidentiality includes cipher algorithm agreement, cipher key agreement, confidentiality of
user data and confidentiality of signaling data. The SEPS-AKA scheme follows the mechanism of
the EPS-AKA protocol and hence is successful with these demands.
6.4 Data Integrity
Data integrity includes integrity algorithm agreement, integrity key agreement, data integrity and
original authentication of signaling data. As illustrated in table I, No one of the scheme presented
in table 1 except the SEPS-AKA scheme attain these purposes since no one of these schemes
provide a level of verification of the IMSI, while in the SEPS-AKA scheme a high level of IMSI
verification is performed using the IMSI extraction as illustrated in equation (9).
7. PERFORMANCE EVALUATION
The evaluation of the performance of the SEPS-AKA scheme is compared to HSK-AKA [7], EC-
AKA [8], SP-AKA [9] and the original EPS-AKA. Therefore, two comparison criteria’s will be
discussed; bandwidth consumption and computation overhead.
7.1.Bandwidth consumption
Measuring the bandwidth consumption requires defining the employed cryptographic algorithm.
Suppose the RSA scheme with 1024-bit key, therefore, the measuring of cipher text size as
following [14]:
o Compute n:
݊ = ∑ ‫ݐݔ݁ݐ݈݊݅ܽ݌‬ ݈݁݊݃‫ݐ‬ℎ ݅݊ ܾ‫ݏ݁ݐݕ‬ (11)
Where n is the length of the transmitted plaintext in bytes.
o Divides plaintext into equal blocks (16 byte)
‫ݏ‬ = ݈ܿ݁݅(
௡
ଵ଺
) (12)
Where S is the integer number of blocks of the plaintext.
o Compute Ciphertext length c_len as :
ܿ_݈݁݊ = ቀ1 +
௙௟௢௢௥(ோௌ஺௞௘௬௦௜௭௘ିଵ)
଼
ቁ ∗ ‫ݏ‬ (13)
Where S is the number of blocks
As shown in the table 2 and figure 3, the SEPS-AKA scheme consumes a bandwidth less than the
EC-AKA, but comparing with SP-AKA, HSK-AKA and original EPS-AKA schemes, SEPS-
AKA consumes a greater bandwidth. Since SP-AKA is not follow the same framework of the
original EPS-AKA. HSK-AKA secures the original AKA based on the pseudonyms approach
without PKI, and there is no security consideration on the original EPS-AKA

Table 2. Bandwidth Consumption
Scheme Total bandwidth in bytes
SEPS-AKA 676
EC-AKA 816
SP-AKA 240
HSK-AKA 336
EPS-AKA 276
Figure 3: Bandwidth Consumption
7.2.Computation Overhead
To measure the computational overhead, we use crypto++5.6.0 benchmark which is compiled
with Microsoft visual C++ 2005 SP1 and runs on Intel core 2 1.83 GHz processor under
WINDOWS VISTA in 32 bit mode [15].
As illustrated in table 3 and figure 4, SEPS-AKA preserves the computational overhead compared
to the EC-AKA scheme, while SP-AKA scheme provide less than computational overhead since,
it is not recognized based on the original framework of the original EPS-AKA scheme. In
addition, the original EPS-AKA and HSK-AKA schemes have no security aspects, therefore,
there is no computational overhead attached with them.
Table 3. Computational overhead
Scheme Computational overhead in microsecond
SEPS-AKA 39540
EC-AKA 89540
SP-AKA 2926
0
200
400
600
800
1000
SEPS-AKA EC-AKA SP-AKA HSK-AKA EPS-AKA
Total bandwidth in bytes

Figure 4: Computational Overhead
8. CONCLUSION
In this paper, we have proposed a secure and efficient EPS-AKA scheme, SEPS-AKA, using PKI
and the RSA scheme computations in order to maintain the problems in the LTE-A authentication
and key management. Compared with other authentication protocols, our proposed scheme
robustly achieves security requirements including; privacy, authentication, confidentiality, and
data integrity. Moreover, as the major contributions of the paper, extensive security analysis
shows that the SEPS-AKA scheme is secure against various malicious attacks such as cyber and
side channel attacks. Furthermore, the SEPS-AKA scheme has a high withstanding to the replay,
DoS, MitM, and Impersonation attacks. The performance evaluation shows that the SEPS-AKA
scheme achieves good bandwidth consumption and less computation overhead.
ACKNOWLEDGEMENTS
This paper is supported in part by the Zamalah Fellowship Program, Gaza, Palestine.
REFERENCES
[1] P. Bhat, S. Nagata, L. Campoy, I. Berberana, T. Derham, G. Liu, X. Shen, P. Zong, and J. Yang,
“LTE-advanced: an operator perspective, “Communications Magazine, IEEE, vol. 50, no. 2, pp. 104–
114, February 2012.
[2] J. Cao, M. Ma, H. Li, Y. Zhang, and Z. Luo, “A survey on security aspects for LTE and LTE-A
networks,” Communications Surveys Tutorials, IEEE, vol. 16, no. 1, pp. 283–302, First 2013.
[3] S. Kanchi, S. Sandilya, D. Bhosale, A. Pitkar, and M. Gondhalekar, “Overview of LTE-A technology,”
in Global High Tech Congress on Electronics (GHTCE), 2013 IEEE, Nov 2013, pp. 195–200.
[4] J.-K. Tsay and S. F. Mjlsnes, “Computational security analysis of the UMTS and LTE authentication
and key agreement protocols,” Report arXiv: 1203.3866v2, Norwegian University of Sciences and
Technology (NTNU), Department of Telematics, Norway, 2013.
[5] Y. Park and T. Park, “A survey of security threats on 4g networks,” in Globecom Workshops, 2007
IEEE, Nov 2007, pp. 1–6.
[6] M. Purkhiabani and A. Salahi, “Enhanced authentication and key agreement procedure of next
generation evolved mobile networks,” in Communication Software and Networks (ICCSN), 2011
IEEE 3rd
International Conference on, May 2011, pp. 557–563.
[7] K. Hamandi, I. Sarji, A. Chehab, I. Elhajj, and A. Kayssi, “Privacy enhanced and computationally
efficient HSK-AKA LTE scheme,” in Advanced Information Networking and Applications Workshops
(WAINA), 2013 27th International Conference on, March 2013, pp. 929–934.
[8] J. Abdo, J. Demerjian, H. Chaouchi, and G. Pujolle, “EC-AKA2 a revolutionary aka protocol,” in
Computer Applications Technology (ICCAT), 2013 International Conference on, Jan 2013, pp. 1–6.
0
20000
40000
60000
80000
100000
SEPS-AKA EC AKA SP AKA
Computational overhead in microsecond

[9] J. B. Abdo, J. Demerjian, K. Ahmad, H. Chaouchi, and G. Pujolle, “EPS mutual authentication and
crypt-analyzing SP-AKA,” in Computing, Management and Telecommunications (ComManTel), 2013
International Conference on, Jan 2013, pp. 303–308.
[10]C. Lai, H. Li, R. Lu, and X. S. Shen, “SE-AKA: A secure and efficient group authentication and key
agreement protocol for LTE networks,” in Computer Networks, Sep 2013, pp. 3492 – 3510.
[11]Y. Zheng, D. He, X. Tang, and H. Wang, “AKA and authorization scheme for 4G mobile networks
based on trusted mobile platform,” in Information, Communications and Signal Processing, 2005 Fifth
International Conference on, 2005, pp. 976–980.
[12]C. Tang, D. Naumann, and S. Wetzel, “Analysis of authentication and key establishment in inter-
generational mobile telephony,” in High Performance Computing and Communications 2013 IEEE
International Conference on Embedded and Ubiquitous Computing (HPCC˙EUC), pp. 1605–1614.
[13] Eli, Atilla and R. Shankaran, “Theory and practice of cryptography solutions for secure information
systems,” in IGI Global, Sep 2013, pp. 1–351.
[14]W. Stallings, “Cryptography and network security: Principles and practice,” in Prentice Hall, Jan 2011,
p. Fifth Edition.
[15]W. Dai, “Crypto++ 5.6.0 benchmarks,” MARCH 2009.
[16]P. Rengaraju, C.-H. Lung, and A. Srinivasan, “Measuring and analyzing WIMAX security and QoS in
testbed experiments,” in Communications (ICC), 2011 IEEE International Conference on, June 2011,
pp. 1–5.
AUTHORS
Zaher Jabr Haddad received the BSc degree from the computer science department,
faculty of applied science, Al-Aqsa University, Palestine in 1999 and MSc degree
from the Department of Information Technology, Faculty of Computers and
Information, Cairo University, Egypt in 2007. He is currently working toward the
Ph.D. degree in Information Technology Department, Faculty of Computers and
Information, Cairo University, Egypt. His interest in wireless network security and
LTE networks
Sanaa Taha received the BSc and MSc degrees from the Department of Information
Technology, Faculty of Computers and Information, Cairo University, Egypt, 2001,
2005, respectively, and the PhD degree in the Electrical and Computer Engineering
from the University of Waterloo, Canada in 2013. She is currently an associated
professor in the Department of Information Technology, Faculty of Computers and
Information, Cairo University, Egypt. Her research interest include wireless network
security, mobile networks security, mobile management, and applied cryptography.
Imane Aly Saroit Ismail received the B.Sc, MSc, and PhD from Department of
Communication, Faculty of Engineering, Cairo University, Egypt, in 1985, 1990, and
1994, respectively. She is currently a full professor in the Department of Information
Technology, Faculty of Computers and Information, Cairo University, Egypt. Her
research interest include wireless network security, mobile networks security, mobile
management, and applied cryptography.

SEPS-AKA: A SECURE EVOLVED PACKET SYSTEM AUTHENTICATION AND KEY AGREEMENT SCHEME FOR LTE-A NETWORKS

More Related Content

What's hot (19)

Similar to SEPS-AKA: A SECURE EVOLVED PACKET SYSTEM AUTHENTICATION AND KEY AGREEMENT SCHEME FOR LTE-A NETWORKS (20)

More from cscpconf (20)

Recently uploaded (20)

SEPS-AKA: A SECURE EVOLVED PACKET SYSTEM AUTHENTICATION AND KEY AGREEMENT SCHEME FOR LTE-A NETWORKS