International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

CONTRIBUTION OF BARKER AND NESTED

BARKER CODE IN CDMA TECHNOLOGY

Sakshi Jain1, Jyotsna V. Ogale2

1
Electronics & Communication Engineering, S.A.T.I, Vidisha (M.P.)-464001, (India)
2
Electronics & Instrumentation Engineering, S.A.T.I, Vidisha (M.P.)-464001, (India)

ABSTRCT

Radar signal processing mainly based on extraction of desired signal from the received signal by rejecting the
unwanted one. During signal transmission signal gets corrupted by noise therefore its detection becomes
difficult. Conventional techniques are not capable to detect the signal where the received power is appreciably
less. Barker and Nested Barker codes are good choice for such application. It is possible to implement these
codes with the help of low cost hardware architecture and FPGA. This paper presents a generation strategy of
barker and nested barker sequence. Comparative performance analysis of both the sequences have been done
on the basis of autocorrelation, cross-correlation, merit factor and bit error rate. It is observed that the nested
barker code gives good rang resolution but provide minimum side lobe.

Keyword - Barker code, nested barker, correlation, bit error rate, merit factor.

I. INTRODUCTION

Most of the radar system and signal processing applications utilize low probability of intercept (LPI) while
keeping the detection capability of radar. LPI signals are designed via pulse compression which is further used
to secure data while communicating. On the other hand in pulse compression to obtain information for distances
higher than distances between transmitted pulses, phase coding is used and for this radar needs short pulses and
for this radar needs short pulses. Pulse compression combines the resolution of short pulse to long pulse of high
energy [1, 2, 3]. In pulse compression transmitted pulse is modulated and correlated with received signal.
Therefore radar applications need suitable codes to maximize the sensitivity and resolution. The best choice for
radar application is Barker code and Nested Barker code [4, 5, 6, 7]. These codes are used as spreading codes.
Barker codes are of limited lengths, maximum length of the code is 13 and PSLR is -22.3 dB [8, 9]. In radar
application, minimum PSLR required is -30dB. By increasing the length of Barker codes, resolution of range
can be improved. In this paper Nested Barker codes are constructed by two Barker codes using product of two
Kronecker functions to increase the length of code. Maximum length is increased to 65. These codes are
compared on the basis of autocorrelation, cross-correlation, and merit factor and bit error rate.
The spreading code helps to retrieve the original data from transmitted one, even if transmitted bits of data got
damaged while transmission. The codes may have different length or different size. IEEE 802.11 standards uses
long codes with code period larger than bit period, and short codes have code period equals the bit period. IEEE

Volume 8, Issue XI, NOVEMBER/2018 Page No:1633

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

802.11b Direct Sequence Spread Spectrum used spreading codes known as Barker code. These codes are
sequences of binary bits of length 11 and 13 [10]. Barker code is shown in Figure: 1.

(a)
Information
Data (011)
Barker code
(10010110101)
DSSS Output

(b)
Information
Data (011)
Barker code
(1111100110101)
DSSS Output

(c)

Figure.1 – (a), (b) and (c) Spreading The Information By Barker Code 11 and 13 bit

II. BARKER CODES

A Barker code is a finite length sequence of binary numbers with the minimum aperiodic autocorrelation. In
pulse compression for radar systems this code results in better range and resolution without shortening a radar
pulse or increase in power. Sequences of Barker are of short length and show good correlation properties and
satisfies the condition |r(ꚍ)| ≤ 1 for (ꚍ≠0) where ꚍ is a delay element[11]. The maximum possible length of
Barker code is 13 which is the ratio of main lobe level of autocorrelation function to the side lobe level of the
autocorrelation function.

Volume 8, Issue XI, NOVEMBER/2018 Page No:1634

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

The Barker sequence x1, x2, x3…..xN is of N length, and its aperiodic Autocorrelation function is given by:

Where K= - (N-1)………+ (N-1), The Integrated Side Lobe Ratio (ISLR) is obtained by dividing the energy
side lobes to the energy of main lobe of its autocorrelation function. As given below:

Peak Side Lobe Ratio (PSLR) is the ratio of absolute maximum among the side lobes to the main peak level in
the Autocorrelation function is given as:

PSLR (dB) =20 log10 (3)

Over PN sequences Barker sequences work well and show several advantages. These sequences not only satisfy
the run and balance property but also have uniform and low side lobes (≤1) in autocorrelation function, but the
size of these families is small. Table.1 lists all known Barker codes.

Table.1- A List of Barker Sequences

Sequence Length of Sequence

2 [1 -1], [1 1]

3 [1 1 -1]

4 [1 -1 1 1], [1 -1 -1 -1]

5 [1 1 1 -1 1]

7 [1 1 1 -1 -1 1 -1]

11 [1 1 1 -1 -1 -1 1 -1 -1 1 -1]

13 [1 1 1 1 1 -1 -1 1 1 -1 1 -1 1]

III. NESTED BARKER CODE

Barker codes show the bi-phase property and with these codes it is possible to achieve smallest side lobes in
autocorrelation function. However, the odd length Barker code is proven to be of length 13 [2] and considered
to be major disadvantage. Barker codes of length 2 and 4 are the only ones of even length. For length 13 code
maximum achievable ratio of a main lobe to peak side lobe is only 22.8 dB which is less than 30dB which is
practically required [12].

Volume 8, Issue XI, NOVEMBER/2018 Page No:1635

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

In the Table given below Nested binary codes are shown. It is found that these codes improve the PSLR
compared to the Barker codes [ 3]. By taking Kronecker product of two Barker codes Nested Barker codes can
be generated with good initial filter response. Let a Barker code of N-bits is and denoted by BN, and a Barker
code of M-bits is and denoted by BM, then an MN bit code can be constructed as BN⊗BM. The Kronecker
product is the BN code repeated M times, with each repetition multiplied by the corresponding element of the
BM code and vice versa Table given below shows the Nested Barker sequences.

Table .2- Nested Barker sequences

S. No. Kronecker product Length of sequence

1 B7 7

2 B11 11

3 B13 13

4 B2 B7 14

5 B4 B4 16

6 B4 B5 20

7 B4 B7 28

8 B7 B7 49

9 B5 B11 55

10 B5 B13 65

IV. GENERATION OF THE CODES

Barker and Nested Barker binary sequence are generated by Linear Feedback Shift Register (LFSR) which
consists of ’n’ number of D type flip-flops. Combination of flip flop produces sequences of 1‟s and 0‟s in a pre
defined manner

Linear Feedback Shift Register (LFSR)

LFSR is a n-bit shift register utilizes the feedback. By shifting the contents of its registers into adjacent register
positions or to a position at the end of the register it usually works. The contents are binary bits. If a shift
register carries the bit pattern 1101 than a single shift to right would result in 0110; another shift yields 0011
[13-14]. To provide the feedback XORing or XNORing is performed over the outs of register pairs. The outputs
of interest of selected stages are referred to as taps by John G. Proakis [13]. Each register is excited by a single
clock pulse externally applied. The taps can be placed anywhere at any stage with XOR or XNOR feedback.

Volume 8, Issue XI, NOVEMBER/2018 Page No:1636

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

Figure 2 shows the structure used to generate random sequence of Length 2n-1 states (n is no. of flip-flops). By
the exhaustive search method the Barker code is selected.

Figure.2 -1:3 bit LFSR

Table. 3- 1:3 bit LFSR sequence

B0 B1 B2 Output
stream
0 1 0
1 0 1 0
1 1 0 1
1 1 0 0
7 Bit Barker
1 1 1 0 Sequence
0 1 1 1
1 0 1 1
1 1 0 1
1 1 1 0 Not
Considerabl
e

Table 3 is an example of LFSR with initial value 010. The output of the feedback is XNOR of B0, B1 and the
output stream.

Input 0 1 1 0 Output

Figure.3- Shift Register

V. PERFORMANCE MEASURE AND It’s ANALYSIS

Correlation Measure - The Mean Square Aperiodic Auto-Correlation (MSAAC) and Mean Square Aperiodic
Cross-Correlation (MSACC) are very useful and widely accepted performance measures to check the
correlation properties of the sequences. These correlation measures are first introduced by Oppermann and

Volume 8, Issue XI, NOVEMBER/2018 Page No:1637

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

Vucetic [15]. Let is a sequence of length N, (n) denotes non-delayed version of (i), and (n+τ) denotes

the delayed version of (j) by ‘τ ‘units. The discrete Aperiodic correlation function is given by:

(4)

MSAAC value for a given code set containing M sequences is defined as:

(5)

(6)

Both the measures are used for comparing the sequence sets. The sequence which shows poor cross-correlation
properties shows good auto-correlation properties, and vice-versa. The sequences with less MSAAC values
means low correlation between the bits within a sample, and sequences with less MSACC values means less the
sample to sample correlation, which makes the information signal less intelligible.

Bit Error Rate (BER) - The BER is a measure of the percentage of bits having less error relative to the total
number of bits. It is expressed as a ratio of number of bits with error & total numbers of bits transmitted. As
defined below:

(7)

BER is the most fundamental measure of system performance [16]. The performance of a system
communicating information digitally is quantified by the probability of bit detection errors in the presence
of thermal noise.

Merit factor - As good cross correlation degrades the auto correlation properties and degradation in auto
correlation directly affects the frequency response of the sequence. Merit factor is calculated to check the
suitability of the frequency response of the sequences in particular application. It is defined as the ratio of
energy in the main lobe of Auto Correlation Function to the total signal energy in side lobes [17]. The Merit
Factor for a sequence, (n), of length N having the auto-correlation function is defined as follow.

= (8)

we may use the inverse of the value calculated for the RAC as the MF.

V. RESULT AND DISCUSSION

Barker and Nested Barker codes for different length are generated and implementation on MATLAB 7.0
version. The performances of these codes are checked by obtaining MSAAC, MSACC, BER and Merit factor
values.

Volume 8, Issue XI, NOVEMBER/2018 Page No:1638

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

Tables. 4 and 5 shows the Aperiodic Auto and Cross correlation for Barker code of length 11 and 13 and
Nested Barker codes of different lengths. In this experiment, efforts are done to maximize auto correlation and
minimize cross-correlation. Cyclic shifts of the sequences are done to test Cross correlations thoroughly. This
function for Barker code is perfect but with the limitation of restricted length. The longest Barker code is only of
13 bit long therefore only serve the purpose of spreading code in CDMA systems for single user. The Barker
code has a very good cross-correlation, with side lobe values less than or equal to 1/N in size and uniform
distribution codes are of limited length when compared to nested barker.

Tables. 6 and 7 shows the merit factor and for Barker code of length 11 and 13 and Nested Barker codes of
different lengths. Both the tables show that merit factor changes with the length of the codes and both the codes
are giving good merit factor. Resolution of the range in radar application can be improved by reducing main
lobe width, side lobe levels and reducing the grating lobes. By suppressing the grating lobes one can increase
the range resolution. In Nested Barker codes grating lobes are suppressed. PSLR is achieved for nested Barker
code of different length ranges from -13.9 dB to -30.8 dB. When compared to the Barker codes PSLR are
improved in nested Barker codes.

Figure (4-10) show plot of MSAAC and MSACC for Barker code of length 11 and 13 and Nested Barker codes
of different lengths. Figure 11 and 12 show the plot of BER versus SNR for Barker code and Nested Barker
codes of different lengths. BER is a powerful method to test the digital transmission system which gives
indication of the performance of the system. Both the codes show BER within limits therefore systems
utilizing these codes will operate satisfactorily.

Table.4- MSAAC and MSACC of Barker Sequence

S. No. Length of Barker Sequence MSAAC MSACC

1 11 0.0826 0. 6463
2 13 0.0710 0.7690

Table.5- MSAAC and MSACC of Nested Barker Sequence

S. No. Length of Nested Barker Sequence MSAAC MSACC

1 14 0.6837 0.5255
2 16 0.6875 0.7188
3 20 0.4500 0.5100
4 49 0.2599 0.2403
5 55 0.2559 0.5556
6 65 0.2424 0.1574

Volume 8, Issue XI, NOVEMBER/2018 Page No:1639

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

Table.6- Measure of Merit Factor of Barker Code Sequence

S.No. Length Of Barker Sequence Proposed Merit Factor Literature PSLR(dB)

MF
1 11 12.10 12.10 -9.5

2 13 13.99 14.08 -12.0

Table.7- Measure of Merit Factor of Nested Barker Code Sequence

S. No. Length Of Proposed Merit Literature PSLR(dB)

Nested Factor MF
Barker
Sequence
1 14 11.46 10.06 -14.0

2 16 11.45 10.48 -20.8

3 20 18.88 10.51 -22.3

4 49 13.84 10.82 -30.8

5 55 13.90 10.72 -13.97

6 65 14.13 11.81 -13.9

(a) (b)

Figure.4 – (a)-(b) Aperiodic Autocorrelation Function of Barker Sequence of length 11 and 13bits

Volume 8, Issue XI, NOVEMBER/2018 Page No:1640

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

(c) (d)

Figure.5 – (c)-(d) Aperiodic Autocorrelation Function of Nested Barker Sequence of length 14 and 16bits

(e) (f)

Figure.6 – (e)-(f) Aperiodic Autocorrelation Function of Nested Barker Sequence of length 20 and 49 bit

(g) (h)

Figure.7 – (g)-(h) Aperiodic Autocorrelation Function of Nested Barker Sequence of length 55 and 65 bits

Volume 8, Issue XI, NOVEMBER/2018 Page No:1641

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

(i) (j)

Figure.8 - (i)-(j ) Aperiodic Cross-correlation Function of Nested Barker Sequence of length 14 and
16 with 11 bits

(k) (l)

Figure.9 - (k)-(l) Aperiodic Cross-correlation Function of Nested Barker Sequence of length 20 and 49 with
11 bit

(m) (n)

Figure.10 - (m)-(n) Aperiodic Cross-correlation Function of Nested Barker Sequence of length 55 and 65 with
11 bits

Figure.11- BER VS SNR of Barker Sequence

Volume 8, Issue XI, NOVEMBER/2018 Page No:1642

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

Figure.12- BER VS SNR of Nested Barker Sequence in Different length

VI. CONCLUSION

These paper discuses the generation of Barker & Nested Barker sequence and its performance parameters.
These codes find enormous significance in data security as used in pulse compression. Particularly suited for
radar application.

VII. ACKNOWLWDGEMENT

The authors would like to thank the Department of Electronics and Communication Engineering, Samrat Ashok
Technological Institute, vidisha (M.P.), for providing the support to carry out this research work.

REFERANCES

[1] N. Levanon, Cross-correlation of long binary signals with longer mismatched filters, IEEE Proc. Radar
Sonar Navigation, vol 152, no.6,pp 377-382, December 2005.
[2] Levanon. N, Mozeson Radar Signal (2004).
[3] ]Merrill I. Skolnik., Introduction To Radar Systems 3/E, (McGraw-Hill, New York).
[4] R. H. Barker, Group synchronizing of binary digital systems, in Communication theory, Butterworth,
London, 1953, pp. 273-287.
[5] Turyn, R., On Barker codes of even length, Proceedings of the IEEE, vol. 51, no. 9, September 1963, p.
1256.
[6] Rao, B. V., and A. A. Deshpande. Why the Barker sequence bit length does not exceed thirteen?. (1988).
[7] Levanon. N, Mozeson, Radar Signal, Wiley, New York, 2004.
[8] E. L. Titlebaum, S. V. Maric and J. R. Belleg. Ambiguity properties of Quadratic congruential Coding,
IEEE Transation on Aerosspace and electronic systems, vol. 27, no. 1 January. 1991, pp. 18-29.
[9] S. R. Gottesman. A class of Pseudo noise-Like pulse compression codes, IEEE Transation on Aerosspace
and electronic systems, vol. 28, no. 2 April. 1992, pp. 355-362.
[10] Jan MIKULKA, Stanislav HANUS, CCK and Barker Coding Implementation in IEEE 802.11b Standard
1-4244-0822-9/07, 2007 IEEE
[11] S. W. Golomb and R. A. Scholtz, Generalized Barker sequences, IEEE Trans. Inform. Theory, vol. IT-11,
no. 4, pp. 533-537, Oct. 1965.

Volume 8, Issue XI, NOVEMBER/2018 Page No:1643

International Journal of Management, Technology And Engineering ISSN NO : 2249-7455

[12] Mark.A.richards, Fundamentals of radar signal processing ( McGraw-hil”l, 2005).

[13] John G. Proakis, Digital Communication (McGraw Hill, Singapore. Pp 502-507,471-475, 2001).
[14] Peterson, William Wesley, and EJ Jr Weldon. Error-correcting codes. The MIT Press, 1972.
[15] I. Oppermann and B. S. Vucetic, Complex spreading sequences with a wide range of correlation properties,
IEEE Trans. Commun., vol. COM- 45, pp. 365-375, March 1997.
[16] Vinay Panwar and Sanjeet Kumar. Bit Error Rate (BER) Analysis of Rayleigh Fading Channels in Mobile
Communication, International Journal of Modern Engineering Research (IJMER) Vol.2, Issue.3, May-June
2012 pp-796-798
[17] M. J. Golay, Sieves for low auto-correlation binary sequences, IEEE Trans, on Information Theory,
23(1):43-51, enero 1977.

Volume 8, Issue XI, NOVEMBER/2018 Page No:1644