Frequency Hopping Spread Spectrum Security Improvement with Encrypted Spreading Codes in a Partial Band Noise Jamming Environment

doi:10.4236/jis.2013.41001

Paper Menu >>

Journal Menu >>

Journal of Information Security, 2013, 4, 1-6

http://dx.doi.org/10.4236/jis.2013.41001 Published Online January 2013 (http://www.scirp.org/journal/jis)

Frequency Hopping Spread Spectrum Security

Improvement with Encrypted Spreading Codes in a

Partial Band Noise Jamming Environment

Amirhossein Ebrahimzadeh, Abolfazl Falahati

Digital Cryptography and Coding Laboratory (DCCS Lab), Department of Electrical Engineering,

Iran University of Science and Technology, Tehran, Iran

Email: [email protected], [email protected]

Received June 17, 2012; revised October 25, 2012; accepted November 7, 2012

ABSTRACT

Frequency Hopping Spread Spectrum (FHSS) system is often deployed to protect wireless communication from jam-

ming or to preclude undesired reception of the signal. Such themes can only be achieved if the jammer or undesired

receiver does not have the knowledge of the spreading code. For this reason, unencrypted M-sequences are a deficient

choice for the spreading code when a high level of security is required. The primary objective of this paper is to analyze

vulnerability of linear feedback shift register (LFSRs) codes. Then, a new method based on encryption algorithm ap-

plied over spreading codes, named hidden frequency hopping is proposed to improve the security of FHSS. The pro-

posed encryption security algorithm is highly reliable, and can be applied to all existing data communication systems

based on spread spectrum techniques. Since the multi-user detection is an inherent characteristic for FHSS, the multi-

user interference must be studied carefully. Hence, a new method called optimum pair “key-input” selection is proposed

which reduces interference below the desired constant threshold.

Keywords: Frequency Hopping Spread Spectrum; Key Distribution Centre; Key Encryption Key; Linear Feedback

Shift Register; Frequency Hopping Code Division Multiple Access; Direct Sequence Spread Spectrum

1. Introduction

Further employment of wireless communication system

to exchange vital and critical electronic information re-

quires an urgent attention to design reliable secure sys-

tems. This requirement is further strengthen for military

communication systems where information transmission

heavily relies upon wireless networks [1].

In fact the major advantage of a mobile set narrow-

band signal transmission is its efficient use of available

frequency due to only a fraction of signal transmission

frequency being used for a single subscriber. Indeed, a

drawback is obvious, as it requires a well coordinated

frequency allocation for different subscribers’ signal

which are now quiet vulnerable to signal jamming and

interception.

In [2,3], the fundamental goal of spread spectrum sys-

tem is considered as; to increase the dimensional charac-

teristic of the signal, hence, to make eavesdropping and/

or jamming more difficult since there are more dimen-

sions of the signal to consider. In fact, the main method

of increasing the dimensionality of the signal is to widen

the signal’s spectral occupancy [2,3].

In spread spectrum techniques, security against tap-

ping and jamming is greater compared with narrowband

spectrum techniques. Signals of spread spectrum are in-

distinguishable from background noise to anyone who

does not know the coding scheme. The disadvantage of

spread spectrum is its relatively high complexity of the

coding mechanism which results in complex radio hard-

ware designs and higher costs. Nonetheless, because of

its remarkable advantages, spread spectrum has been

adopted by many wireless technologies. For example, the

IEEE 802.11b standard for wireless LAN employs DSSS

over the 2.4-GHz free spectrum, whereas the Bluetooth

standard uses frequency hopping spread spectrum (FHSS)

for simplicity [1-3].

Furthermore, the main weakness of the wireless com-

munication security system is the simplicity of accessing

the communicating signal through the channel. Eaves-

droppers can easily place an antenna in the desired field

and after demodulation, the message bits can be obtained

in the base-band form. If the messages are encrypted,

after storing the encrypted messages with some crypto-

analysis methods, the original message can be smeared

out. Now, if the received radio signals from the wireless

channel is spread in a form that the intruder cannot ac-

A. EBRAHIMZADEH, A. FALAHATI

cess the despread spectrum and receives only a signal

similar to noise, a perfectly secure radio transmission

channel is achieved [3-5].

Moreover, specifically the security of the frequency

hopping code division multiple access (FH-CDMA) sys-

tem mainly relies on the long-code generator that con-

sists of a 42-bit long-code mask generated by a 42-bit

LFSRs. However, if eavesdroppers can obtain 42 bits of

plaintext-cipher-text pairs, the long-code mask can be

recovered after dropping the transmission on the traffic

channel for about one second [3,6,7].

The fast correlation attack method based on a recently

established linear statistical weakness of decimated

LFSR sequences for reconstruction of LFSR code is de-

scribed in [8]. With this method eavesdropper can re-

cover LFSR sequence that he knows the LFSR feedback

polynomial. A method of blind estimation of PN code in

multipath fading direct sequence spread spectrum sys-

tems is proposed in [9]. In this article a combed method

is presented to estimate the unknown PN spreading se-

quence for direct sequence spread spectrum (DS-SS)

signals in frequency selective fading channel. It is proven

that LFSR codes are vulnerable to cipher-text-only at-

tacks [10] and security weakness of white Gaussian se-

quence is investigated in [11].

This preface and further studies show that LFSR codes,

white Gaussian sequences and other unencrypted codes

have security weaknesses and can be recovered by eaves-

droppers. So a method which can guarantee systems

against the probable attacks is urgently required.

In this manuscript, a new method called hidden fre-

quency hopping spread spectrum is proposed to augment

the built-in security of FH-CDMA systems by applying

cryptographic algorithm in the channelization code se-

quence.

2. Security Enhancement in FHSS System

with Encryption Hidden within SS

In FHSS technique, several users spread their signal

spectrum through available wideband frequency spec-

trum as narrowband sections with a special code which is

called frequency hopping. These codes must have a low

cross-correlation since other signals have little interfere-

ence over the desired signal. On the other hand, although

M-sequences which are generated by LFSR have fair

cross-correlation properties but they produce a weak se-

curity system for eavesdroppers to track the transmitted

spread signals. Therefore, FH-CDMA uses a long-code

to scramble the signal in wireless channels, thus the se-

curity is set up in the physical layer. The available secu-

rity which is produced by this method is very low and not

suitable for data communication considered. In this arti-

cle, for security enhancement, a model is proposed that

every user encrypts a special spreading code (e.g. a code

that is made by the M-sequence generators) with his pri-

vate key. The model is shown in Figure 1. Encrypted

codes are then used as the spreading code in the chan-

nelization section. At destination, the receiver who knows

his private key is able to regenerate the spreading code to

de-spread the transmitted signal [3].

On the other hand, the security by the proposed me-

thod is related to the encryption algorithm, not to the

LFSR complexity. If a suitable algorithm such as RC5,

IDEA or any block cipher algorithm is chosen, then a

desired high privacy can be obtained [12,13].

3. The Proposed System Model

Although spread spectrum systems are used for narrow-

band interference mitigation and have good efficiency in

preventing intentional and unintentional channel inter-

ference, if jammer uses similar spreading codes method,

it can be successful in deteriorating such techniques. The

level of signal destruction depends on similarity between

jammer and transceiver PN codes. This mechanism is

different for FHSS and DSSS systems but FH systems

are desired. In this method, jammer operates intelligently,

after accessing the channel and receive spread signals, it

finds spreading technique and PN sequence pattern. Then

it generates similar PN pattern and can synchronize itself

with the transceiver system to track the modulation type.

It should be mentioned that jammer can be located be-

tween transmitter and receiver so to provide the man in

the middle attack. So jammer can interfere with data sig-

nal or change receiver to a useless one and mask itself as

an allowable user.

A proposed hidden frequency hopping method can be

used to prevent sequence pattern disclosure. Therefore,

complexity in this process solely depends upon encrypt-

tion complexity. Let’s consider MFSK transceiver which

employs FHSS with encrypted PN sequences, Gaussian

noise power and partial band noise jamming function j(t).

Suppose that jammer can access channel and obtain de-

sired information from this system.

First a Key Distribution Centre (KDC) generates and

transmits agreeable session key to receiver by secure

procedure. Session key is a symmetrical key that BTS

and SS know and its transmission would be performed by

asymmetric pair public-privet key encryption. In this

manner symmetric key encrypts SS public key and only

SS can decrypt it. Then desired key is transmitted trough

unsecure channel by secure process. Public asymmetric

key is called Key Encryption Key (KEK). After key ex-

change, transmitter and receiver have the same encryp-

tion key to be able to encrypt PN codes that generate

hidden narrow band frequencies. The transceiver system

can now be synchronized, track encrypted PN codes,

access hidden hopped frequencies and finally obtains

A. EBRAHIMZADEH, A. FALAHATI





ˆk

















Figure 1. Propose d F HSS with hidden PN sequenc e s for MFSK receiver (M = 2).

original data signals.

Therefore, the jammer applies interference power on

narrow band hopped channels randomly and so the bit

error rate can be computed as

 

s-encrypted

no-hit

Perror hitPPerror no-hitP 

sent-signalH s

TTt

sent-signalH s

TTt

Shit S

(1)

where PS presents symbol error probability and Phit is the

probability of signal hitting the jammer.

Suppose that the transceiver and jammer have the

same hop period TH, the number of hoped channel NH

with different start signaling. Because jammer doesn’t

have session key, it can’t obtain spreading codes, syn-

chronize with transceiver and to know start transceiver

signaling. Thus, the time of transceiver signaling is given

(2)

where ts0 is the transceiver signalling start time. The jam-

mer signalling time is

(3)

where tj0 is start time of jammer signaling, and 0



& 0

. So jammer and transceiver signal hit occurs

in joint time hop period. This value can be expressed as

tT



ent-jammersent-signal s

min T,T



 (4)

Figure 2 describes the transceiver signal and jammer

signal collision behaviour. So within this behaviour, the

probability of jammer and transceiver signal hit in k-th

hop is

hit





no-hit hit

P1P

(5)

and

(6)





If jammer uses partial band noise jammer, with stan-

dard M-ary FSK modulation that uses one out of M fre-

quencies each second, the bit error probability can be

obtained as



 



s-encrypted

2N2 N

1e1e

T2M1

1e1e

T2M1















 

 

 















(7)

where

E is bit energy, 0

is the one-sided noise

spectral density, and T

is the total AWGN and jam-

mer noise.

Considering α is maximized by as

ttt





maxsent-signalsent-jammer ttt

min T,T







(8)

sent-ja mmer

Figure 2. Collision of jammer and transceiver signals.

A. EBRAHIMZADEH, A. FALAHATI

And maximum term for synchronization Sync



can be

given by

0. 14

Syncmax H



 (9)

In fact, this is obtained when jammer can synchronize

itself with victim transceiver system. If jammer doesn’t

have session key, it can’t access max



The cross-correlation of some codes such as Gold and

Kassami is lower than the encrypted codes, so a method

must be used to optimize the input interference of the

system to an acceptable extent. For this reason, an inter-

ference threshold is selected for the channel according to

the channel interference average for 100 or 1000 times

tests. In these tests the channel minimum value interfer-

ence are calculated and a constant threshold is selected

for a channel with a number of users, then the multi-user

interference is estimated for each user who enters the

network and this value is compared with the threshold

level. If the result is less than threshold level, the opti-

mum pair “key-input” is saved for new user and the data

is sent confidently such that the interference do not ex-

ceed the threshold level. If the interference value is more

than the threshold level, the user has to generate another

PN and give it to the cryptographer for generating a fre-

quency whose interference is not more than the threshold

level. If p is probability of failure and (1 − p) is the prob-

ability of success, the probability of achieving the desired

code after k-tries is



 

p1 p

 (10)

4. Simulation Results

To perform simulation purposes, an averaging over dif-

ferent keys is employed to mitigate the dependency of

BER results on the chosen keys. The simulation results

given in Figure 3 show that the mean value of interfer-

ence among the encryption algorithm outputs is higher

than the M-sequence codes. These values are estimated

for 256 channel hops and 1000 iterations. It is dependent

upon the selected keys and the encryption algorithm in-

puts. Figure 4 represents the number of users trying to

find optimum pair “key-input” for 32, 64, 128, 256 and

512 channel hops.

Figure 5 indicates the BER performance when the

number of users is increased. This means that the inter-

ference is directly proportional to number of users. Fig-

ure 6 represents the effect of FHSS system to mitigate

the interference in multi-user channel for 64, 128, 256

and 512 channel hops. Figure 7 compares m-sequence

codes with encrypted codes considering both optimum

and non-optimum key method in order to select pair in-

put-key for 256 channel hop. This performance shows

that the interference value for encrypted codes is higher

05 10 15 20 2530

0. 02

0. 04

0. 06

0. 08

0. 1

0. 12

Users

M ean of I nterferenc e

M-s equence codes

Encry pt ed c odes

Figure 3. Comparing mean of interference M-sequence and

encrypted codes in channel with 256 hop.

1800

05 10 15 20 2530

200

400

600

800

1000

1200

1400

1600

Number of Us ers

Mean of T ry

32 Hop

64 Hop

128 Hop

256 Hop

512 Hop

Figure 4. Calculation of mean try to access desired pair

input-key with optimum interference.

100

10-1

0 1 2 3 45 6 7 89

10-5

10-4

10-3

10-2

probabl it y of error

Eb/N0(dB)

user=1

user=20

user=10

Probability of error

user=5

Figure 5. Probability of error for various number of users

with 256 channel hop.

A. EBRAHIMZADEH, A. FALAHATI 5

0 2 4 6 81012 14

10-6

10-5

10-4

10-3

10-2

10-1

Eb/N0(dB)

probabli t y of error

bpsk

mult i-us er Interferenc e

64 Hop

128 H op

256 H op

512 H op

Probability of erro

-1

-2

Figure 6. Probability of error for different channel hops in

AWGN channel.

0 1 2 3 45 6 7 89

-5

-4

-3

-2

-1

Eb/No(dB)

probablity of error

bpsk

64 Hop

128 Hop

256 Hop

512 Hop

Probability of error

Figure 7. Comparing probability of error for M-sequence,

encrypted and optimum encrypted code in 256 channe l hop.

than unencrypted m-sequence codes and employing op-

timum method improves this value further.

Figure 8 represents the probability of decoding when

wrong code is used. By comparing Figure 6 with Figure

8, it’s clear that if wrong codes are used, bit error rate

becomes very high. In other words, if eavesdropper uses

wrong code or key, it receives partial error frequency. To

receive hidden frequencies, it must have access to a ses-

sion key that victim transceiver employs.

5. Conclusions

A thorough investigation revealed security weaknesses of

PN sequences. These drawbacks are challenged deploy-

ing a new method called hidden frequency hopping to

augment the built-in security of FHSS systems. It is

named hidden frequency hopping method since a known

cryptographic algorithm is hidden within the PN codes.

0 1 2 3 45 6 789

-5

-4

-3

Eb/No(dB)

probablity of error

M-s equence codes

E nc ry pted codes without opt i m um key

Probability of error

E nc ry pted codes with opt i m um k ey

Figure 8. Uncertainty performance when a wrong key is in-

serted.

The jammer and transceiver signal hit occurrence in

joint time hop period, the parameter α, is determined by

analysis through the jammer and transceiver signalling

start times. This value allows determination of the prob-

ability of the jamming signal colliding with the desired

signal. It is found that, maximum value of α is reached

when the jammer can determine the signalling start time

to synchronize itself with the victim (a common un-en-

crypted method) and it is observed that with the proposed

technique, the jammer can never reach maximum value

to synchronize itself with the victim.

Simulation results show that when the proposed en-

crypted codes are utilized, the received channel interfere-

ence increases. Naturally, this phenomenon makes the

signal detection procedure more complex. Therefore, an

optimum pair “key-input” algorithm is proposed to re-

duce the associated interference to the desired level. For

instance, by employing codes with a good orthogonal

behaviour, indeed, still algorithm can provide small

amount of error but it also reduce the data transmission

speed.

Furthermore, the performance of the encrypted FH/SS

code is as good as unencrypted sequences when correct

key is applied. When a wrong key is employed, system

security is guaranteed.

6. Acknowledgements

Authors would like to thank the financial support of

ITRC (Iran Telecommunication Research Company) for

this project.

REFERENCES

[1] P. Zheng, L. Peterson, B. Davie and A. Farrel, “Wireless

A. EBRAHIMZADEH, A. FALAHATI

Network Complete,” Elsevier publisher, Amsterdam, 2009.

[2] Telecommunications Industry Association, “Mobile Sta-

tion-Base Station Compatibility Standard for Dual-Mode

Wideband Spread Spectrum Cellular System,” Telecom-

munications Industry Association, Arlington, 1998.

[3] M. Tafaroji and A. Falahati, “Improving Code Division

Multiple Access Security by Applying Encryption Me-

thods over the Spreading Codes,” IET Communication,

Vol. 1, No. 3, 2007, pp. 398-404.

doi:10.1049/iet-com:20060295

[4] J. L. Massey, “Shift-Register Synthesis and BCH Decod-

ing,” IEEE Transactions on Information Theory, Vol. 15,

No. 1, 1969, pp. 122-127. doi:10.1109/TIT.1969.1054260

[5] I. Mansour, G. Ghalhoub and A. Quilliot, “Security Ar-

chitecture for Wireless Sensor Networks Using Frequency

Hopping and Public Key Management,” IEEE Interna-

tional Conference on Networking, Sensing and Control

(ICNSC), Delft,11-13 April 2011, pp. 526-531.

[6] V. K. Gray, “IS-95 CDMA and CDMA2000,” Pren-

tice-Hall, Upper Saddle River, 2000.

[7] J. K. Tugnait and T. Li, “Blind Detection of Asynchro-

nous CDMA Signals in Multipath Channels Using Code-

Constrained Inverse Filter Criteria,” IEEE Transactions

on Signal Process, Vol. 49, No. 7, 2001, pp. 1300-1309.

doi:10.1109/78.928685

[8] J. Colic, “Towards Fast Correlation Attacks on Irregularly

Clocked Shift Registers, Information Security Research

Centre,” EUROCRYPT’95 Proceedings of the 14th An-

nual International Conference on Theory and Application

of Cryptographic Techniques, Louvain-la-Neuve, pp. 248-

262.

[9] X. Xu, “Blind Estimation of PN Code in Multipath Fad-

ing Direct Sequence Spread Spectrum Systems,” 11th

IEEE International Conference on Communication Tech-

nology Proceeding, Hangzhou, 10-12 November 2008, pp.

213-216.

[10] Q. Ling, T. Li and J. Ren, “Physical Layer Built-in Secu-

rity Enhancement of DS-CDMA Systems Using Secure

Block Interleaving,” Conference Record of the Thirty-

Eighth Asilomar Conference on Signals, Systems and Com-

puters, Pacific Grove, 7-10 November 2004, pp. 1105-

1109.

[11] L. Gang and A. Nakansu and M. Ramkumar, “Security

and Synchronization in Watermark Sequences,” 2002

IEEE International Conference on Acoustics, Speech, and

Signal Processing (ICASSP), Orlando, 13-17 May 2002,

pp. IV-3736-IV-3739.

doi:10.1109/ICASSP.2002.5745468

[12] B. Schneier, “Applied Cryptography: Protocols, Algo-

rithms and Source Code in C,” 2nd Edition, John Wiley

& Sons, Hoboken, 1996.

[13] B. S. Kaliski, and L. Y. Yiqun, “On the Security of the

RC5 Encryption Algorithm,” RSA Laboratories Techni-

cal Report, Cambridge, 1998.