Channel equalization using watermark as a training sequence

Adv. Radio Sci., 9, 179–185, 2011 www.adv-radio-sci.net/9/179/2011/ doi:10.5194/ars-9-179-2011 © Author(s) 2011. CC Attribution 3.0 License.

Advances in Radio Science

Channel equalization using watermark as a training sequence M. K. Samee and J. G¨otze Information Processing Lab, Dortmund University of Technology, Otto-Hahn-Str. 4, 44227 Dortmund, Germany

Abstract. In this paper blind equalization algorithms based on digital watermarking are presented. Training sequence is sent along with the data in the form of watermark. Robust CDMA based watermarking algorithm is used to watermark the data. Watermark, which is spread over the data, does not consume extra bandwidth. So with the help of extracted watermark channel can be equalized just by using simple mean square algorithm. Two algorithms for equalization are proposed in this paper. In algorithm I the receiver does not require to know training sequence in advance. Using algorithm II, the sender does not require to send training sequence either, and data can still be equalized using simple LMS algorithm at low computational complexity.

1

Introduction

Intersymbol Interference (ISI) is a very common type of distortion which a signal undergoes during transmission (Proakis, 2000). Trained equalization is a technique widely used to cope with ISI. A training sequence is sent over the channel and with the help of received and already known training sequence channel is equalized. Traditional trained equalization scheme works with simple algorithms like Least Mean Squares (LMS) (Haykin, 1996). Most crucial drawback of traditional trained equalization is that it consumes extra bandwidth. To save the bandwidth, some blind equalization techniques, which do not require training sequences, are used to equalize channels. These blind equalization schemes are usually based on computationally complex algorithms like Constant Modulus Algorithm (CMA) (Litwin, 1999). So the cost of saving bandwidth is increased computational complexity. Correspondence to: M. K. Samee ([email protected])

Digital Watermarking is a well known technique used for authentication, copyrights protection and applications like avoiding illicit copying etc. Watermark is simply a sequence of some bits which is hidden in host data (e.g. videos, images and someother multimedia data). If some meaningful watermark is embedded in host data watermarking can be used for some other purposes as well. Tracing watermarks are used for blind quality assessment for multimedia communication in (Campisi, 2003). A multipurpose public-key cryptosystem based on image watermarking is developed in (Ding, 2008). End-to-end QoS provision and control in wireless communication by means of digital watermarking is presented in (Benedetto, 2006). Digital watermarking is used for dynamic and adaptive equalization in (Maity, 2007; Neto, 2006) respectively. However in (Maity, 2007; Neto, 2006) receiver should know the watermark (training sequence) in advance, which is not required in proposed scheme. In comparison to blind equalization, traditional trained equalization is known to be computationally efficient. But the most crucial drawback of trained equalization is that it consumes extra bandwidth. Another disadvantage is that the receiver must have prior information of the original training sequence. There are some blind equalization algorithms, where training sequences are not used and data can still be equalized (Litwin, 1999; Zarzoso, 2005). To design a blind equalization algorithm convergence of the algorithm and computational complexity are two important issues. Constant Modulus Algorithm (CMA) (Litwin, 1999) is one of well known algorithm used for blind equalization. Disadvantage for CMA is its slow convergence. Therefore, researchers tried to lessen the computational complexity of CMA and tried to make convergence faster (Sun, 2007; Dabeer, 2003). In this work, a completely different approach to blind equalization is taken by using digital watermarking. The proposed method does not require training sequences to be known in advance and works with standard trained equalization algorithms (e.g. LMS) at low computational complexity.

Published by Copernicus Publications on behalf of the URSI Landesausschuss in der Bundesrepublik Deutschland e.V.

180

M. K. Samee and J. G¨otze: Channel equalization using watermark as a training sequence

Furthermore, proposed scheme can simultaneously be used for blind equalization and usual watermarking applications (Podilchuk, 2001). Proposed schemes can be looked from two different viewpoints. First, how watermarking can be used for blind equalization (the receiver is not required to know the training sequence). Second, how digital watermarking can be used to correct errors in transmitted images. Additionally the proposed algorithms can simultaneously be used for blind equalization and usual watermarking applications. In this paper, two watermarking based equalization algorithms are presented, for two different scenarios: 1. when channel specific training sequence or specific watermark is required, 2. when watermark/training sequence can be any sequence (can be a part of data). In algorithm I, training sequence is embedded in the data in the form of watermark. This training sequence is also sent over the channel. On receiving side, the watermark is extracted and with the help of received training sequence and extracted watermark channel can be equalized. In this algorithm receiver does not required to know the training sequence in advance. Consider the following scenario; An image (data) distributor distributes watermarked images (data) containing different watermarks to several clients. Distributor also sends the watermark along with the watermarked image (data) through the channel. On the receiving sides, receivers only have a watermark extractor (all the receivers use the same extracting software). Therefore, the receivers can extract the watermark and use it as a reference training sequence. The watermark which distributor has sent through the channel can be treated as received training sequence. Now, the channel can be estimated and hence received, erroneous, watermarked images can be corrected. In algorithm II, a chunk of data, which is a part of the data actually to be transmitted through the channel, is used as a watermark. This chunk is embedded in the data as a watermark. With the help of received chunk and extracted chunk (extracted watermark) the channel can be equalized blindly. Watermarking algorithm used for these blind equalization schemes, must have the following two qualities: – Robust (to withstand distortions due to transmission). – Carry enough payload (which can be used as training sequence). CDMA based spread spectrum watermarking scheme (Samee, 2007) fulfills the above requirements. The main advantage of spread spectrum watermarking is that each watermark bit is embedded in a number of pixels. Because of that it is proved to be robust in transmission. By using multiple orthogonal spreading codes, CDMA based watermarking Adv. Radio Sci., 9, 179–185, 2011

scheme can carry more payload (large watermark). Normalized Least Mean Square (NLMS) algorithm (Haykin, 1996) is used in this scheme for equalization. Other equalization algorithms e.g. Least Mean Square (LMS) and Recursive Least Square (RLS) can also be used. This paper is organized as follows: Sect. 2 presents an overview of previously described CDMA based watermarking algorithm. In Sect. 3 blind equalization scheme using digital watermarking is described. Section 4 presents the experimental results and Sect. 5 concludes this paper. 2

Overview of the watermarking scheme

An oblivious watermarking scheme is presented in (Samee, 2007), which is further improved by using by-part interleaving in (Samee, 2008). In this watermarking scheme every bit of the watermark is hidden in mutually similar frequency coefficients in the DCT domain. These specially selected frequency coefficients are preferably in middle frequency range. Every vector of the modified frequency coefficients ij0 is formed according to: ij0 = ij + α[b1 s1 + b2 s2 + ... + bk sk ]

(1)

where ij is a vector of similar frequency coefficients, bi is a watermark bit, α is the gain factor and si are the spreading codes. Watermark bits can be extracted by using: bi = sign < ij0 ,si >

if |ij · siT | < |αsi · siT |

(2)

For complete description of the watermarking scheme see (Samee, 2007). This scheme is proved to be robust against many intentional and unintentional attacks. Transmission of watermarked image through a channel can be seen as an unintentional attack. The watermark easily survives the transmission attack by using this watermarking scheme. The use of multiple spreading codes makes this scheme able to carry large watermark. 3

Watermarking-based blind equalization

Figure 1 shows how trained equalization works. In order to equalize the data receiver must know the reference training sequence in advance. Error e(n) is calculated with the help of received training sequence and reference training sequence. Now by minimizing e(n), the inverse of the channel is estimated and weights (taps) w(n) are updated. The received data can be equalized by using w(n). Figure 2 shows algorithm I for blind equalization, where a training sequence is hidden (or superimposed) in the data by using digital watermarking. After transmission, if this hidden training sequence is extracted from the received data without much errors, the extracted training sequence can be used as reference training sequence (which is supposed to be provided by the sender). www.adv-radio-sci.net/9/179/2011/

Watermark as aandTraining Sequence ee: Channel Equalization Using Watermark as aasTraining Sequence Received M. K. Samee J. G¨otze: Channel equalization using watermark a training sequence Received

Received 181watermarked image

training sequence

data

Training Sequence

Data

Equalizer

w(n)

Watermarke

Watermarked image

Adaptive control algorithm

e(n)

Channel

d(n)

+

−

Equa Wate extr

Chann

Channel

+

: Channel Equalization Using Watermark as a Training Sequence Received

Received training sequence

Training Sequence

Datadata

Corrected data

ed quence

Equalizer

Channel raditional trained equalization technique. e(n) d(n)

e(n)Received d(n)

Received training sequence

data

+

Equalizer

w(n)

+

−

Watermarked image

Received watermarked image

Adaptive control algorithm

w(n)

3 Received selected part

Equalizer/ Watermark extractor

+

e(n)

e(n)

Received selected part Adaptive control algorithm

w(n)

d(n)

w(n

3. Algorithm II for b

w(n) Fig.

Equalizer/ Watermark extractor

Correct watermarke

Equalizer/ Watermark extractor

Adaptive control algorithm

Channel

Received watermarked image

+

Adaptive control algorithm

Corrected data

Received watermarked image

e(n)

d(n)

−

+

Corrected + watermarked image

Extracted ce training sequence (which is supposed to be prowatermark Fig. 1. Traditional trained equalization technique. Corrected y1. the sender). Traditional trained equalization technique. watermarked image Fig. 3. Algorithm II for blind equ +

−

−

+

+

d(n)

Corrected watermarked image

Corrected data

+

Extracted watermark

Fig. 3. Algorithm II for blind equalization.

Traditional trained equalization technique. Watermark chnique. Watermarked image Fig. 3. Algorithm for blindequalization. equalization. Fig. 3. Algorithm II forIIblind

rence training sequence (which is supposed to be pronce training sequence (which is supposed to be prod by the sender).

by the sender).

Watermarked image

Received watermarked image

ceived ermark

chunk Fig. Selected 4. Selecting a chu data. Chunk spread over the stream of data

Received watermark

Received watermarked image

Equalizer/ Watermark extractor

Equalizer/ Received Watermark watermarked w(n) extractor

image

Channel Adaptive control Received w(n) algorithm watermark

Adaptive control algorithm

dat

Initialization. If prio data. Chunk spread over the stream of data e(n) w(n) ˆ is available, us Received Initialization. If prior knowledge on the tap-weight vector Figure 3 shows algorithm II for blind equalization, where watermark + w(n) w(0). ˆ Otherwise, selected chunk use of data is hidden (or in the forset w ˆ a is available, it data to select ansuperimposed) appropriate value Fig. 4.− Selecting a chunk of and hidding it in the s Adaptive 4 ). stream ofset data w(0). ˆ entire Otherwise, w(0) ˆ(Fig.= 0.After transmission, this hid-

e(n)

Equalizer/ Watermark d(n) extractor Corrected Corrected watermarked image

Selected chunk Chunk spread over t

4. Selecting a chunkofofdata data and it in of Fig. 4.Fig.Selecting a chunk andhidding hidding it the in stream the stream of data. data. Fig. 4. Selecting a chunk of

Channel

rk

Stream of data

Watermark

Watermark

Chunk

Stream of data

Channel

s supposedWatermarked to be image pro-

Stream

Extracted watermark

−

w(n)

+

d(n) control

+

data.

+ algorithm

Extracted watermark

e(n)

watermarked image

Algorithm IFig. for2.blind equalization. d(n) Algorithm I for blind equalization.

−

+ +

Initialization. If prior knowle w(n) ˆ Data is available, use it to s : u(n) w(0). ˆ (a)Given Otherwise, set w(0) ˆ : M-b =

den chunk of data is extracted from the received data. This extracted chunk can be used as a reference training sequence Data and the received chunk can be used as a received training se(a)Given : u(n) M-by-1 at time n quence. Thus :e(n) can betap-input calculatedvector and NLMS algorithm can be d(n) used to: desired update w(n). Hence received data can be response at time n equalized blindly.

d(n) : des

Extracted watermark

Initialization. If prior :knowledge on ofthe tap-weigh lgorithm I forCorrected blind equalization. (b)To be computed w(n ˆ Data + 1) = estimate tap-weight For equalization Normalized LMS algorithm is used. This e(n) watermarked image algorithm works Haykin (1996): (a)Given : an u(n) :n+1 M-by-1 tap vector at time w(n) ˆ where is use asitfollows to select appropriate va 3 shows algorithm II for blind equalization, a available, (b)To be computed : w ˆ ed chunk of data is hidden (or superimposed) in the enComputation : nw(0) = 0,1,2,... d(n) : desired resp + www.adv-radio-sci.net/9/179/2011/ w(0). ˆ Otherwise, set ˆ = 0. Adv. Radio Sci., 9, 179–185, 2011 eam of data (Fig. 4 ). After transmission, this hidden

2. Algorithm I for blind equalization. −

e(n) = d(n) − w + extracted data is from the II received data. This ex3ofshows algorithm for blind equalization, whereˆa (n)u(n) µ ˜

d chunk can be used as a reference training sequence

H

w(n ˆ + 1) = w(n) ˆ +

∗ (b)To be computed : w(n ˆ (3)+ 1) u(n)e (n)

fading c maximu M. K. Samee and J. G¨otze: Channel equalization using watermark as a training sequence channel 4: Frequen channel Table 1. BER in extracted watermarks using algorithm I. Doppler Received images Equalized images channels Lena Milk drop Lena Milk drop Selected part equal to % % % %

atermarked image can simultaneously be used for blind zation as well as for usual watermarking applications. 182

Original Image

1 2 3 4

8.2 16.8 0.39 2.54

8.59 16.8 10.16 10.55

0 0 0 0

0 0 0 0

Computation : n = 0,1,2,... Dummy Watermarked

Dummy Watermarked Dummy watermarked chunk is spread over remaining image

e(n) = d(n) − wˆ H (n)u(n) µ˜ ˆ + 1) = w(n) ˆ w(n + u(n)e∗ (n) a+ k u(n) k2

(3)

Normalized LMS is a modified version of LMS algorithm. LMS algorithm also works fine in proposed blind equalization method. However simulations have shown that Normalized LMS algorithm works slightly better than LMS algorithm. LMS algorithm experiences a gradient noise amplification problem when u(n) is large, because the correction ˆ applied to the tap-weight vector w(n) at iteration n + 1 is directly proportional to the tap-input vector u(n). In normalized LMS algorithm correction applied to tap-weight vector ˆ w(n) at iteration n + 1 is normalized with respect to squared Euclidean norm of tap-input vector u(n) at iteration n. That is why it is called “normalized” LMS.

(a) Lena image.

Watermarked Image

Fig. 6. Original Lena and M

Fig. 5. Forming a watermarked image, considering implementation

3.1

Implementation problem

Formingproblem. a watermarked image, considering implementation 4.1 Algorithm I A very crucial problem arose while implementing proposed m. scheme using digital watermarking. A part of an image is P arameter : M = number of taps a = positive constant µ˜ = adaption constant (step-size parameter) 0 < µ˜ < 2 Initialization: if prior knowledge on the tap-weight vector ˆ w(n) is available, use it to select an appropriate value for ˆ ˆ w(0). Otherwise, set w(0) = 0. Data: (a) Given : u(n) : M-by-1 tap-input vector at time n d(n) : desired response at time n

ˆ + 1) = estimate of tap-weight (b) To be computed : w(n vector at time n+1 Adv. Radio Sci., 9, 179–185, 2011

selected, and later it is spread over the whole image forming watermarked image. Obviously, watermarked image is slightly different from the original image. When this watermarked image is transmitted, channel can not be equalized. Because, the extracted watermark is a part of the original image and received part is from watermarked image. Hence both of them are different and the channel can not be equalized using extracted watermark and received part of the watermarked image. One of the simplest solutions to this problem is to hide the selected part in the remaining image (other than selected part). However, this simple solution causes problems for usual watermarking applications, like copyrights protection etc. From general watermarking applications point of view, it is not good to leave a reference to the original image. Watermark should be spread over the whole image. If reference to the original image is present in the watermarked image, some denoising algorithm can detect watermark as noise. So watermark can be removed easily. Therefore, this scheme can not be used for usual watermarking applications.

First of all, training sequ watermark. 512 bits lon

www.adv-radio-sci.net/9/179/2011/

M. K. Samee and J. G¨otze: Channel equalization using watermark as a training sequence Table 2. Lena received and equalized using algorithm I.

channels 1 2 3 4

PSNR Received Equalized dB dB 29.0929 28.9833 22.0178 13.8834

77.5595 69.6869 Infinite 42.9292

Table 4. BER in extracted watermarks using algorithm II.

Erroneous bits Received Equalized % % 8.94 12.2 33.58 9.29

1 2 3 4

PSNR Received Equalized dB dB 29.2870 29.1856 23.0038 13.8618

74.1472 92.3162 Infinite 21.8373

Erroneous bits Received Equalized % % 8.51 11.65 33.58 9.29

3.2

1 2 3 4

channels

www.adv-radio-sci.net/9/179/2011/

0 0 0 0

0 0 0 0

2552 dB MSE

87.1311 86.5184 102.3162 88.8920

Erroneous bits Received Equalized % % 9.67 13.04 33.60 8.06

0.00 0.00 0.00 0.00

(4)

Using MATLAB, these images are transmitted over four different channels: channel 1: channel 2: channel 3:

channel 4:

4.1

In all the experiments 512×512 (8 bits/pixel, gray scale) images (Lena Fig. 6a and Milk drop Fig. 6b) are used. 512 bits long watermark is hidden in 3rd, 4th, 5th and 6th rows of 8×8 blocks of DCT coefficients (Samee, 2007). Two spreading codes, each of length 512, are used to spread the watermark bits. All the images are watermarked at 40 dB (here α = 0.01). The quality of an image is measured in terms of Peak Signal to Noise Ratio (PSNR).

26.8461 26.8677 22.5406 12.3285

PSNR = 10log

Possible security issue

Experimental results

8.93 10.52 4.37 17.26

PSNR Received Equalized dB dB

1 2 3 4

As watermark is a part of the watermarked image, it can be considered as a possible security issue. By using proposed scheme, watermark can be any part of watermarked image, which is very difficult to locate for an evedropper. Furthermore, because of proposed implementation scheme noise level (watermark intensity) is same over the whole image. So it is difficult to distinguish between selected part, watermark, and remaining watermarked image. 4

4.76 6.75 0.2 24.40

Equalized images Lena Milk drop % %

Table 5. Lena received and equalized using algorithm II.

0.03 0.00 0.00 1.27

This problem is solved by watermarking selected portion with some dummy bits first and then hide this dummy watermarked selected portion into remaining image (Fig. 5). Now this watermarked image has same level of noise over whole image and has no reference to the original image. Therefore, this watermarked image can simultaneously be used for blind equalization as well as for usual watermarking applications.

Received images Lena Milk drop % %

channels

0.01 0.09 0.00 0.00

Table 3. Milk drop received and equalized using algorithm I.

channels

183

{0.986, 0.845, 0.237, 0.123+0.310i} {0.986, 0.845, 0.537, 0.323, 0.123} Frequency-flat (single path) Rayleigh fading channel, sample time 1 × 10−5 and maximum Doppler shift 0.09 Hz. Frequency-flat (single path) Rician fading channel, sample time 1 × 10−5 , maximum Doppler shift 0.99 Hz and Rician factor equal to 1.

Algorithm I

First of all, training sequence is embedded in an image as watermark. 512 bits long training sequence (watermark) is sent over the channel followed by the watermarked image. On the receiving side, watermark is extracted from the watermarked image. The extracted watermark has a few erroneous bits (table 1) but the BER is good enough to use it as a reference training sequence. Normalized LMS with 8 ˆ weights, µ˜ = 0.01 and w(0) = 0 are used. The calculated ˆ weights w(n) are used to equalize the received image. Received and corresponding equalized Lena images are shown in Fig. 7. Errors and PSNR values of received and equalized Adv. Radio Sci., 9, 179–185, 2011

ency communication signals. Furthermore, the suitabilfferent watermarking schemes forfor thethe proposed blind different watermarking schemes proposed blind channel 1: {0.986, 0.845, 0.237, 0.123+0.310i} channel 1: {0.986, 0.845, 0.237, 0.123+0.310i} tion algorithm can be studied. ences will be done to develop a similar scheme for radio hnces will be done to develop a similar scheme for radio zation algorithm can be studied. Samee: Equalization Using Watermark Training Sequence Samee: Equalization Using Watermark Training Sequence channel {0.986, 0.845, 0.537, 0.323, 0.123} 1842: M. K.suitabilSamee andChannel J.Channel G¨otze: Channel equalization using watermarkas asasaa a training sequence channel 2: {0.986, 0.845, 0.537, 0.323, 0.123} cy communication signals. Furthermore, the ency communication signals. Furthermore, the suitabilProakis, Digital Communications, fourth edition, McGraw G. Proakis, Digital Communications, fourth edition, McGraw fferent watermarking for blind different watermarking schemes forthe theproposed proposed blind channel 3: Frequency-flat (single path) Rayleigh channel 3: schemes Frequency-flat (single path) Rayleigh (a)(a)Lena (b) Lenaimage. image. (b)Watermarked WatermarkedLena. Lena. 2000. ces , 2000. ences −5 Equalization −5 Samee: Channel Using Watermark as a Training Sequence will done develop a similar scheme for radio Samee: Channel Equalization Using Watermark as a Training Sequence hation willbe be done develop a similar scheme for radio algorithm can be studied. Tableto6.to Milk drop received and equalized using algorithm II. zation algorithm can be studied. fading channel, sample time 1 × 10 and and fading channel, sample time 1 × 10 Haykin: Theory, third edition, Prentice Hall, Haykin:Adaptive AdaptiveFilter Filter Theory, third edition, Prentice Hall, cy communication signals. Furthermore, the suitabilency communication signals. Furthermore, the suitabilmaximum Doppler shift 0.09 bits Hz. Hz. Proakis, Digital Communications, fourth edition, McGraw PSNR Erroneous maximum Doppler shift 0.09 6.Proakis, G. Digital Communications, fourth edition, McGraw (a)(a)Lena image. (b)(b)Watermarked fferent watermarking schemes for proposed blind watermarking schemes forthe the proposed blind Lena image. WatermarkedLena. Lena. will be done to develop a similar scheme for radio h.different will be done to develop a similar scheme for radio channels Received Equalized Received Equalized 000. Litwin, Jr.: Blind Channel Equalization, IEEE Potentials, R. Litwin, Jr.: Blind Channel Equalization, IEEE Potentials, , 2000. channel 4: Frequency-flat (single path) Rician fading channel 4: Frequency-flat (single path) Rician fading nces ences dB dB % % ation algorithm can be studied. zation algorithm can beTheory, studied. cy communication signals. Furthermore, the suitabilency communication signals. Furthermore, the suitabilaykin: Adaptive Filter Theory, third edition, Prentice Hall, me 18, Issue 4,4,Page(s):9-12, October-November 1999. −5 −5 ume 18, Issue Page(s):9-12, October-November 1999. Haykin: Adaptive Filter third edition, Prentice Hall, channel, sample time 1 × 10 , maximum channel, sample time 1blind ×0.00 10 , maximum 1 Carli, 27.2704 83.6830 4.56 ifferent watermarking schemes for the proposed blind different watermarking schemes for the proposed Campisi, Marco Gaeteno Giunta and Alessandro o Campisi, Marco Carli, Gaeteno Giunta and Alessandro 6. fourth edition, McGraw G.Proakis, Proakis,Digital DigitalCommunications, Communications, fourth edition, McGraw Doppler shift 0.99 Hz4.76 and Rician factorfactor Doppler shift 0.99 HzComand Rician 2can 27.3037 74.7575 0.03 (a)(a)Lena (b)(b)Watermarked Lenaimage. image. WatermarkedLena. Lena. Litwin, Jr.: Blind Channel Equalization, IEEE Potentials, Blind Quality Assessment System for ation algorithm bebe studied. ization algorithm can studied. Blind Quality Assessment System forMultimedia Multimedia ComR. Litwin, Jr.: Blind Channel Equalization, IEEE Potentials, 2000. ,i: 2000. 3 equal 22.6232 102.3162 10.32 0.00 nces ences to 1. equal to 1. e 18, Issue 4, Page(s):9-12, October-November 1999. cations Using Tracing Watermarking, IEEE Transactions nications Using Tracing Watermarking, IEEE Transactions

ume 18,Adaptive Issue 4, Page(s):9-12, October-November 1999. Haykin: third edition, Hall, Haykin: Adaptive FilterTheory, Theory, third edition,Prentice Prentice Hall, 4Filter 13.2914 102.3162 14.88 0.00 Campisi, Marco Carli, Gaeteno Giunta and Alessandro gnal Processing [see also IEEE Transactions on AcousSignal Processing [see also IEEE Transactions on Acouso Campisi, Marco Carli, Gaeteno Giunta and Alessandro 6. G.Proakis, Proakis,Digital DigitalCommunications, Communications,fourth fourthedition, edition,McGraw McGraw nces ences Blind Quality Assessment System for Multimedia Comand Signal Processing], Volume 51, Issue 4,4, ,i:Speech, Speech, and Signal Processing], Volume 51, Issue (a)(a)Lena (b)(b)Watermarked Lenaimage. image. WatermarkedLena. Lena. Blind Quality Assessment System for Multimedia ComJr.: Blind Channel Equalization, IEEE Potentials, R. Litwin, Jr.: Blind Channel Equalization, IEEE Potentials, 2000. , Litwin, 2000. s):996-1002, April 2003. ations Using Tracing Watermarking, IEEE Transactions e(s):996-1002, April 2003. nications Using Tracing Watermarking, IEEE Transactions me 18, 4,4,Page(s):9-12, October-November 1999. ume 18,Issue Issue Page(s):9-12, October-November 1999. (c)(c)Received through Equalized image (a) Lena image. (b) Watermarked Lena.for Haykin: Adaptive Filter third edition, Prentice Hall, Received throughchannel channel(d) (d) Equalized image for(c). (c). Haykin: Adaptive FilterTheory, Theory, third edition, Prentice Hall, G. Proakis, Digital Communications, fourth edition, McGraw Proakis, Digital Communications, fourth edition, McGraw Ding, Zi Lin and Li Wang: A Multipurpose Publicgnal Processing [see also IEEE Transactions on Acousen Ding, Zi Lin and Li Wang: A Multipurpose PublicSignal Processing [see also IEEE Transactions on AcousCampisi, o6. Campisi,Marco MarcoCarli, Carli,Gaeteno GaetenoGiunta Giuntaand andAlessandro Alessandro (a) Lena image. (b) Watermarked Lena. (a) Lena image. (b) Watermarked Lena. 1. 1. , 2000. 2000. Cryptosystem Based Image Watermarking, 4th International Speech, and Signal Processing], Volume 51, Issue 4, 4, Based Image Watermarking, 4th International Speech, and Signal Processing], Volume 51, Issue Blind Assessment System for Comi:Cryptosystem BlindQuality Quality Assessment System forMultimedia Multimedia ComLitwin, Jr.: Blind Channel Equalization, IEEE Potentials, R. Litwin, Jr.: Blind Channel Equalization, IEEE Potentials, Haykin: Adaptive Filter Theory,third thirdedition, edition, Prentice Hall, Haykin: Adaptive Filter Theory, Prentice Hall, erence on Wireless Communications, Networking and Mo):996-1002, April 2003. nference on Wireless Communications, Networking and Moe(s):996-1002, April 2003. cations Using Tracing Watermarking, IEEE nications Using Tracing Watermarking, IEEETransactions Transactions (c)(c)Received me 18, 4,4,Page(s):9-12, October-November 1999. ume 18,Issue Issue Page(s):9-12, October-November 1999. Receivedthrough throughchannel channel(d)(d)Equalized Equalizedimage imageforfor(c). (c). 6. Computing, WiCOM ’08., Page(s):1-4, 12-14 Oct. 2008. Ding, Zi Lin and Li Wang: A Multipurpose PublicComputing, WiCOM ’08., Page(s):1-4, 12-14 Oct. 2008. n Ding, Zi Lin and Li Wang: A Multipurpose PublicProcessing [see also IEEE onAlessandro Signal Processing [see also IEEETransactions Transactions onAcousAcous1. Campisi, Marco Carli, Gaeteno Giunta and Alessandro o.gnal Campisi, Marco Carli, Gaeteno Giunta and 1. R. Litwin, Jr.: Blind Channel Equalization, IEEE Potentials, Litwin, Jr.: Based Blind Channel Equalization, IEEE Potentials, co Benedetto, Giunta and Neri: Endryptosystem Image Watermarking, 4th International esco Benedetto, Gaetano Giunta andAlessandro Alessandro Neri: EndCryptosystem Based Image Watermarking, 4th International Speech, and Signal Processing], Volume 51, Issue 4,4, Speech, andGaetano Signal Processing], Volume 51, Issue Blind Quality Assessment System for Comi: Blind Quality Assessment System forMultimedia Multimedia Comume 18, Issue 4, Page(s):9-12, October-November 1999. me 18, Issue 4, Page(s):9-12, October-November 1999. d QoS Provision and Control in Wireless Communication rence on Wireless Communications, Networking and MoEnd QoS Provision and Control in Wireless Communication ference on Wireless Communications, Networking and Mos):996-1002, April 2003. e(s):996-1002, April 2003. Using Tracing Watermarking, IEEE Transactions nications Using Tracing Watermarking, IEEE Transactions (c)(c)Received Receivedthrough throughchannel channel(d)(d)Equalized Equalizedimage imageforfor(c). (c). oations Campisi, Marco Carli, Gaeteno Giunta and Alessandro Campisi, Marco Carli, Gaeteno Giunta and Alessandro ms bybyMeans ofof Digital Watermarking Signal Processing, tems Means Digital Watermarking Signal Processing, omputing, WiCOM ’08., Page(s):1-4, 12-14 Oct. 2008. WiCOM ’08., Page(s):1-4, 12-14 Oct. 2008. Ding, Zi Lin and Li Wang: ATransactions Multipurpose nComputing, Ding, Zi Lin and Lialso Wang: ATransactions Multipurpose Publicgnal Processing [see also IEEE ononPublicAcousSignal Processing [see IEEE Acous1. 1. i: Blind Quality Assessment System for Multimedia ComBlind Quality Assessment System for Multimedia Comnternational Symposium on Communication SysInternational Symposium onWireless Wireless Communication Sysosco Benedetto, Gaetano Giunta andand Alessandro Neri: EndBenedetto, Gaetano Giunta Alessandro Neri: Endryptosystem Based Image Watermarking, 4th International Cryptosystem Based Image Watermarking, 4th International (a) Lena image. (b)Issue Milk drop. Speech, and Signal Processing], Volume 51, 4,4, Speech, and Signal Processing], Volume 51, Issue nications Using Tracing Watermarking, IEEE Transactions cations Using Tracing Watermarking, IEEE Transactions ISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. s. ISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. d QoS Provision and Control in Wireless Communication (c) Received through channel 1. (d) Equalized image for (c). End QoS Provision and Control in Wireless Communication erence onon Wireless Communications, Networking and Moference Wireless Communications, Networking and Mos):996-1002, April 2003. e(s):996-1002, April 2003. (a) Lena image. (b) Milk drop. (a) Lena image. (b) Milk Signal Processing [see also IEEE Transactions on Acous- drop. gnal Processing [see also IEEE Transactions on AcousMaity, Malay K. Kundu and Seba Maity: An Efficient (c) Received (c) Receivedthrough throughchannel channel(d)(d)Equalized Equalizedimage imageforfor(c). (c). P. Maity, Malay K. Kundu and Seba Maity: An Efficient Fig. 6. Original Lena and Milk drop images. ms by Means of Digital Watermarking Signal Processing, ems by Means of Digital Watermarking Signal Processing, ’08., Page(s):1-4, 12-14 Computing, WiCOM ’08., Page(s):1-4, 12-14Oct. Oct.2008. 2008. Ding, Lin and Li Wang: A AMultipurpose Publicnomputing, Ding,Ziand ZiWiCOM Lin and LiProcessing], Wang: Multipurpose Public,ital Speech, and Signal Volume 51,Issue Issue Speech, Signal Processing], Volume 51, 4,4, al Watermarking Scheme for Dynamic Estimation of Wire1. Watermarking Scheme for Dynamic Estimation of Wire1. ternational Symposium on Wireless Communication SysInternational Symposium on Wireless Communication Sysco Benedetto, Gaetano Giunta and Neri: Endsco Benedetto, Gaetano Giunta andAlessandro Alessandro Neri: Endryptosystem Based Image Watermarking, 4th International Cryptosystem Based Image Watermarking, 4th International images for both Lena and Milkdrop are shown in Tables 2 e(s):996-1002, April 2003. s):996-1002, April 2003. Channel Condition, International Conference onon ComputChannel Condition, International Conference ComputISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. ISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. (f)(d) Receivedthrough throughchannel (f)Equalized Equalizedimage imagefor for(e). (e). (c)Received channel (d) (c). (c)(e) (c). ds. QoS Provision and Control in Wireless Communication End QoS Provision and Control in Wireless Communication rence on Wireless Communications, Networking and Moference on Wireless Communications, Networking and Mo- show (e) and 3, respectively. Here, second and third columns en Ding, Zi Lin and Wang: AMultipurpose Multipurpose PublicDing, Zi Lin and LiLi Wang: Adrop PublicTheory and Applications. ICCTA ’07, Page(s):671-675, 5-7 Theory and Applications. ICCTA ’07, Page(s):671-675, 5-7 Maity, Malay K. Kundu and Seba Maity: An Efficient P. Maity, Malay K. Kundu and Seba Maity: An Efficient Fig. 6. Original Lena and Milk drop images. Fig. 6. Original Lena and Milk images. 2. 2. 1. ms bybyMeans ofofDigital Signal Processing, ems Means Digital Watermarking Signal Processing, PSNR value ofWatermarking received and 12-14 equalized images with respect 1. omputing, WiCOM ’08., Page(s):1-4, Oct. 2008. Computing, WiCOM ’08., Page(s):1-4, 12-14 Oct. 2008. Cryptosystem Based Image Watermarking, 4th International Cryptosystem Based Image Watermarking, 4th International hital 2007. rch 2007. lInternational Watermarking Scheme for Dynamic Estimation of of WireWatermarking Scheme for Dynamic Estimation Wirenternational Symposium on Wireless Communication SysSymposium on Wireless Communication Systo watermarked images. Fourth and fifth columns show pero Benedetto, Gaetano Giunta Neri: Endsco Benedetto, Gaetano Giuntaand andAlessandro Alessandro Neri: Endnference onWireless Wireless Communications, Networking and Moerence on Communications, Networking and MoUliani Neto, Leandro de Jo˜ ao2006. T.T. RoUliani Neto, Leandro deC.C.T.T.Gomes, Gomes, Jo˜ aoMarcos Marcos Ro-images. (e)(e)Received hannel Condition, International Conference on ComputChannel Condition, International Conference on Computcentage of erroneous bits in received and equalized ISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. s. ISWCS ’06. Page(s):655-659, 6-8 Sept. Receivedthrough throughchannel channel(f)(f)Equalized Equalizedimage imageforfor(e). (e). dComputing, QoS and Control ininWireless Communication End QoSProvision Provision and Control Wireless Communication WiCOM ’08., Page(s):1-4, 12-14 Oct. 2008. Computing, WiCOM ’08., Page(s):1-4, 12-14 Oct. 2008. and Madeleine Bonnet: Adaptive Equalization based on no and Madeleine Bonnet: Adaptive Equalization based on heory and Applications. ICCTA ’07, Page(s):671-675, 5-7 Theory and Applications. ICCTA ’07, Page(s):671-675, 5-7 Column 4 and 5 of Table 1 show the BER in the watermarks P.ems Maity, Malay K. Kundu andSeba SebaMaity: Maity: An Efficient Maity, Malay K. Kundu and An Efficient 2. 2. ms byBenedetto, Means of Digital Watermarking Signal Processing, by Means of Digital Watermarking Signal Processing, esco Gaetano Giunta and Alessandro Neri: Endco Benedetto, Gaetano and Alessandro End4.1 Algorithm Ifor .1 Algorithm I Giunta rmarking, 2006 International Telecommunications Sympoermarking, 2006 International Telecommunications Sympo2007. extracted from equalized images. These Neri: watermarks are error ch 2007. ital Watermarking Scheme Dynamic Estimation of Wireal Watermarking Scheme for Dynamic Estimation of Wirenternational Symposium ononWireless Communication SysSymposium Wireless Communication Sys(e) Received through channel 2. (f) Equalized image for (e). End QoS Provision and Control inWireless Wireless dInternational QoS Provision Control inGomes, Page(s):743-748, 3-6 Sept. 2006. m, Page(s):743-748, 3-6 Sept. 2006. free andand easily be for usual watermarking applications. iani Neto, Leandro deInternational C. T.used Gomes, Jo˜aJo˜ oaCommunication Marcos T. T. RoUliani Neto, Leandro de C. T. oCommunication Marcos RoChannel Condition, Conference ComputChannel Condition, International Conference ononComputISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. s. ISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. (e) Receivedthrough throughchannel channel (f)(f)Equalized Equalizedimage imageforfor(e). (e). (e) Received tems by Means of Digital Watermarking Signal Processing, ms byMadeleine Means of Digital Watermarking Signal Processing, mad Kashif Samee and uJ¨ rgen G¨ o’07, tze: Increased Robustmmad Kashif Samee andJ¨ uAdaptive rgensequence G¨ otze: Increased RobustFirst of all, training isAn embedded an2.2.image and Bonnet: Adaptive Equalization based on5-7 First of all, training sequence isPage(s):671-675, embedded in as as no and Madeleine Bonnet: Equalization based onaninimage Theory and Applications. ICCTA Theory and Applications. ICCTA ’07, Page(s):671-675, 5-7 Maity, Malay K. Kundu and Seba Maity: Efficient P. Maity, Malay K. Kundu and Seba Maity: An Efficient International Symposium Wireless Communication Sysnternational Symposium onon Communication Sys4.2 Algorithm IIWireless and Security of Digital Using DS-CDMA, and Security of Digital Watermarking Using DS-CDMA, marking, 2006 International Telecommunications Sympoermarking, 2006 International Telecommunications Sympoch2007. 2007. hsital watermark. 512Watermarking bits long training sequence (watermark) watermark. 512 bits long training sequence (watermark) is is al Watermarking Scheme for Estimation ofof WireWatermarking Scheme forDynamic Dynamic Estimation Wires. ISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. ISWCS ’06. Page(s):655-659, 6-8 Sept. 2006. edings ofofthe 7th International Symposium onon Signal ceedings theLeandro 7thIEEE IEEE International Symposium Signal Page(s):743-748, 3-6 Sept. 2006. m, Page(s):743-748, 3-6 Sept. 2006. Uliani Neto, de C. T. Gomes, Jo˜ a o Marcos T. Roliani Neto, Leandro de C. T. Gomes, Jo˜ a o Marcos T. RoChannel Condition, International Conference on ComputChannel Condition, International Conference on ComputFirst eight rows of and anSeba image are watermarked with eight (e)(e)Received P. Maity, Malay K.and Kundu Seba Maity: An Efficient Receivedthrough throughchannel channel (f)(f)Equalized Equalizedimage imageforfor(e). (e). Maity, Malay K. Kundu Maity: An Efficient ssing and Information Technology, Page(s):189-193, Dec. cessing and Information Page(s):189-193, Dec. mad Kashif Samee J¨uTechnology, rgen G¨ oG¨ tze: Increased Robustmmad Kashif Samee and J¨ uand rgen oEqualization tze: Increased Robustno and Madeleine Bonnet: Adaptive Equalization based ondummy and Madeleine Bonnet: Adaptive based on Theory and Applications. ICCTA ’07, Page(s):671-675, 5-7 Theory and Applications. ICCTA ’07, Page(s):671-675, 5-7 dummy bits. First 504 bits are selected from the ital Watermarking Scheme for Dynamic Estimation of Wire2. 2. al Watermarking Scheme for Dynamic Estimation of Wire7. nd Security of of Digital Watermarking Using sermarking, and Security Digital Watermarking UsingDS-CDMA, DS-CDMA, 2006 International Telecommunications Sympomarking, 2006 International Telecommunications Sympowatermarked part and hidden in remaining rows of the imch 2007. hmad 2007. Channel Condition, International Conference ComputChannel Condition, International Conference onon Comput(g)(g) (g). Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and (g). mmad Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and (e) Receivedthrough throughchannel channel(h) (f)Equalized Equalizedimage imagefor for(e). (e). (f)(h) ceedings of the 7th IEEE International Symposium on Signal edings of the 7th IEEE International Symposium on Signal m, Page(s):743-748, 3-6 Sept. 2006. Page(s):743-748, 3-6 Sept. 2006. age. These watermarked images (watermarked at 40 dB) are (e) Received Uliani Neto, Leandro C.T. T.Gomes, Gomes, Jo˜ Marcos T.RoRoliani Neto, Leandro dedeC.ICCTA Jo˜ aoaoMarcos T. Theory and Applications. ICCTA ’07, Page(s):671-675, 5-7 Theory and Applications. ’07, Page(s):671-675, 5-7 3. ng Pai: Digital Watermarking: Spreading Code versus 3. Ting Pai: Digital Watermarking: Spreading Code versus 2. 2. cessing and Information Technology, Page(s):189-193, Dec. sing and Information Technology, Page(s):189-193, Dec. transmitted over the same four channels, mentioned above. mmad Kashif Samee and J¨ uAdaptive rgen oEqualization tze: Increased Robustmad Kashif Samee and J¨ uAdaptive rgen G¨oG¨ tze: Increased Robustno and Madeleine Bonnet: Equalization based and Madeleine Bonnet: based onon rch 2007. hnel 2007. Coding, Proceedings of the 3rd IEEE International annel Coding, Proceedings of the 3rd IEEE International 7. (g) Received through channel 3. (h) Equalized image for (g). On the receiving side, watermark is extracted from the wasand andSecurity Security DigitalWatermarking Watermarking UsingDS-CDMA, DS-CDMA, ofof Digital Using ermarking, 2006 International Telecommunications Sympomarking, 2006 International Telecommunications SympoUlianiNeto, Neto, Leandro de C.T.Control T.Control Gomes, Jo˜ aoMarcos Marcos T.RoRoUliani Leandro dergen C. Gomes, Jo˜ aextracted oSignal T. osium on Communications, and Signal Processing, mposium on Communications, and Processing, (g) Receivedthrough throughchannel channel(h)(h)Equalized Equalizedimage imageforfor(g). (g). mmad Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and (g) Received termarked image. The watermark from the received mad Kashif Samee, J¨ u G¨ o tze, Shanq-Jang Ruan and ceedings 7thIEEE IEEE International SymposiumononSignal Signal edings ofofthethe7th International Symposium m, Page(s):743-748, 3-6 Sept. 2006. Equalization Page(s):743-748, 3-6 Sept. 2006. no and Madeleine Bonnet: Adaptive Equalization based on and Madeleine Bonnet: Adaptive based on , Page(s):1409-1412, March 2008. lta, Page(s):1409-1412, March 2008. image, which forms the training sequence, has a few erroTing Pai: Digital Watermarking: Spreading Codeversus versus 3. 3. ng Pai: Digital Watermarking: Spreading Code cessing and Information Technology, Page(s):189-193, Dec. ssing and Information Technology, Dec. mmad Kashif Samee and urgen oPage(s):189-193, tze: Increased Robustmad Kashif Samee and J¨ uJ¨ rgen G¨oG¨ tze: Increased Robustermarking, 2006 International Telecommunications Sympormarking, 2006 International Telecommunications SympoZarzoso and Pierre Comon: Blind and Semi-Blind Equale Zarzoso and Pierre Comon: Blind and Semi-Blind Equalneous bits (column 2 and 3 of Table 4). The received selected nnel Coding, Proceedings of the 3rd IEEE International el Coding, Proceedings of the 3rd IEEE International 7.andSecurity sand Security ofConstant Digital Watermarking Using DS-CDMA, of Digital Watermarking Using DS-CDMA, m, Page(s):743-748, 3-6Sept. Sept. 2006. 3-6 2006. ˆ part is treated asPower the received training sequence. w(n) are nPage(s):743-748, Based on the Criterion, IEEE Transacion Based on the Constant Power Criterion, IEEE Transacmposium on Communications, Control and Signal Processing, osium on Communications, Control and Signal Processing, Receivedthrough throughchannel channel(h)(h)Equalized Equalizedimage imageforfor(g). (g). (g)(g)Received mmad Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and mad Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and ceedings of the 7th IEEE International Symposium on Signal edings of the 7th IEEE International Symposium on Signal mmad Kashif Samee and J¨ u rgen G¨ o tze: Increased Robustcalculated using Normalized LMS with M = 8 weights, step mad Kashif Samee and J¨urgen G¨ oIssue tze: Increased Robuston Signal Processing, Volume 53, 11, sPage(s):1409-1412, onPage(s):1409-1412, Signal Processing, Volume 53, Issue 11,Page(s):4363Page(s):4363ta, March 2008. March 2008. 3. 3. Ting Pai: Digital Watermarking: Spreading Code versus ng Pai: Digital Watermarking: Spreading Code versus cessing andInformation Information Technology, Page(s):189-193, Dec. and Dec. ˆ ˆ size µ˜ofDigital =Digital 0.01Technology, and w(0) = 0.Page(s):189-193, The calculated weights w(n) are sssing and Security Watermarking UsingDS-CDMA, DS-CDMA, Security ofPierre Watermarking Using November 2005. 5, November 2005. eand Zarzoso Comon: Blind and Semi-Blind EqualZarzoso andand Pierre Comon: Blind and Semi-Blind Equalnnel Coding, Proceedings of the3rd 3rd IEEE International nel Coding, Proceedings of the IEEE International 7. used to equalize the received image. Errors and PSNR values ceedings ofthe the 7thIEEE IEEE International Symposium Signal edings ofChao 7th International Symposium onon Signal un and Chao Zhao: Optimizing Blind Equalization IntelliSun and Zhao: Optimizing Blind Equalization Intelliion Based on the Constant Power Criterion, IEEE Transacnmmad Based on the Constant Power Criterion, IEEE Transacmposium on Communications, Control and Signal Processing, osium on Communications, Control and Signal Processing, of received and corresponding equalized images are shown in (g)(g)Received Receivedthrough throughchannel channel(h)(h)Equalized Equalizedimage imageforfor(g). (g). Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and mad Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and cessing and Information Technology, Page(s):189-193, Dec. ssing and Information Technology, Page(s):189-193, Dec. Algorithm forProcessing, Communication Systems, 3rd ts,ta,Page(s):1409-1412, Algorithm forWireless Wireless Communication Systems, 3rdInterInteron Signal Volume 53, Issue 11, Page(s):4363on Signal Processing, Volume 53, Issue 11, Page(s):4363Page(s):1409-1412, March 2008. March 2008. Tables 5 and 6. Here, second and third columns show PSNR 3. 3. Ting Pai: Digital Watermarking: Spreading Code versus ng Pai: Digital Watermarking: Spreading Code versus 7. Conference (i) Received through channel 4. (j) Equalized image for (i). nal on Computation, ICNC Volonal Conference onNatural Natural Computation, ICNC2007, 2007, Vol- to wa5, November 2005. November 2005. value of received and equalized images with respect eZarzoso Zarzoso and Pierre Comon: Blind and Semi-Blind Equaland Pierre Comon: Blind and Semi-Blind Equalnnel Coding, Proceedings of the 3rd IEEE International nel Coding, Proceedings of the 3rd IEEE International (i) Received through channel (j) Equalized (i) Received through (j) (i). (g) Received channel (h) Equalizedimage imagefor for(i). (g). mmad Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and (g) Received through (h) (g). mad Kashif Samee, J¨ u rgen G¨ o tze, Shanq-Jang Ruan and 3, Page(s):146-149, 24-27 August 2007. eSun 3, Page(s):146-149, 24-27 August 2007. and Chao Zhao: Optimizing Blind Equalization Intellitermarked images. Fourth and fifth columns show percentage nion and Chao Zhao: Optimizing Blind Equalization IntelliFig. 7. Lena received and equalized. Based on the Constant Power Criterion, IEEE TransacBased on the Constant Power Criterion, IEEE Transacmposium on Communications, Control and Signal Processing, osium on Communications, Control and Signal Processing, 4. 4. 3. Ting Pai: Digital Watermarking: Spreading Code versus 3. ing Pai: Digital Watermarking: Spreading Code versus Dabeer and Elias Masry: Convergence Analysis of3rd ConDabeer and Elias Masry: Convergence Analysis ofthe the Conoffor erroneous bits in received and equalized images. Column ton Algorithm Wireless Communication Systems, 3rd Interlgorithm for Wireless Communication Systems, Inters,ta, onSignal Signal Processing, Volume 53,Issue Issue 11,Page(s):4363Page(s):4363Processing, Volume 53, 11, Page(s):1409-1412, March 2008. Page(s):1409-1412, March 2008. annel Coding, Proceedings of the 3rd IEEE International nel Coding, Proceedings of the 3rd IEEE International Modulus Algorithm, IEEE Transactions on Information nt Modulus Algorithm, IEEE Transactions on Information 4 show the BER in the watermarks extracted 4 and 5 of Table onal Conference on Natural Computation, ICNC 2007, Volal Conference on Natural Computation, ICNC 2007, Vol5, November 2005. November 2005. eZarzoso Zarzoso and Pierre Comon: Blindand and Semi-Blind EqualPierre Comon: Blind Semi-Blind Equalmposium on Communications, Control and Signal Processing, posium onand Communications, Control and Signal Processing, Receivedthrough throughchannel channel (j)(j)Equalized Equalizedimage imageforfor(i). (i). from equalized images. ory, Volume 49, Issue 6, Page(s):1447-1464, June 2003. ry, Volume 49, Issue 6, Page(s):1447-1464, June 2003. (i)(i)Received e 3, Page(s):146-149, 24-27 August 2007. ,nPage(s):146-149, 24-27 August 2007. Sun andChao Chao Zhao: Optimizing Blind Equalization Intelliun and Zhao: Optimizing Blind Equalization Intelliion Based the Constant Power Criterion, IEEETransacTransacBased ononthe Constant Power Criterion, IEEE Fig. 7. Lena received and equalized. Fig. 7. Lena received and equalized. lta,Page(s):1409-1412, Page(s):1409-1412, March 2008. a, March 2008. 4. 4. Dabeer Elias Masry: Convergence Analysis the Conabeer andand Elias Masry: Convergence Analysis of of the Conton Algorithm for Wireless Communication Systems, 3rd InterAlgorithm for Wireless Communication Systems, 3rd InterseZarzoso on Signal Processing, Volume 53,Issue Issue 11,Page(s):4363Page(s):4363Signal Processing, Volume 53, 11, Zarzoso and Pierre Comon: Blind and Semi-Blind Equaland Pierre Comon: Blind and Semi-Blind Equalt Modulus Algorithm, IEEE Transactions onInformation Information Adv. Radio Sci., 9,Transactions 179–185, 2011 www.adv-radio-sci.net/9/179/2011/ Modulus Algorithm, IEEE on onal Conference on Natural Computation, ICNC 2007, Volnal Conference on Natural Computation, ICNC 2007, Vol5, November 2005. 2005. ion Based the Constant Power Criterion, IEEE TransacnNovember Based onon the Constant Power Criterion, IEEE Transac(i) Received through channel Equalizedimage imageforfor(i). (i). (i) Received through channel (j)(j)Equalized ory, Volume 49, Issue 6, Page(s):1447-1464, June 2003. Volume 49,Zhao: Issue 6, Page(s):1447-1464, June 2003. ey, Page(s):146-149, 24-27 August 2007. 3, August 2007. Sun and Chao Zhao:24-27 Optimizing Blind Equalization Intelli- Fig. and Chao Optimizing Blind Equalization Intellisn3,Page(s):146-149, on Signal Processing, Volume 53, Issue 11,Page(s):4363Page(s):4363on Signal Processing, Volume 53, Issue 11, Fig. 7. Lena received and equalized. 7. Lena received and equalized. 4. 4. Dabeer and Elias Masry: ConvergenceAnalysis Analysis ofthe the Conabeer and Elias Masry: Convergence of3rd Cont5, Algorithm for Wireless Communication Systems, 3rd InterAlgorithm for Wireless Communication Systems, InterNovember 2005. November 2005.

M. K. Samee and J. G¨otze: Channel equalization using watermark as a training sequence 5

Conclusions

How watermarking can be used to equalize transmission channels only using the received data is discussed in this paper. Simulations have shown that proposed schemes can correct almost all the errors from received watermarked images. An important advantage of this scheme is that it works at the complexity of traditional trained equalization methods unlike other very complex blind equalization methods. Proposed schemes can simultaneously be used for blind equalization as well as for usual watermarking applications. Further research will be done to develop a similar scheme for radio frequency communication signals. Furthermore, the suitability of different watermarking schemes for the proposed blind equalization algorithm can be studied. References John G. Proakis: Digital Communications, fourth edition, McGraw Hill, 2000. Simon Haykin: Adaptive Filter Theory, third edition, Prentice Hall, 1996. Litwin Jr., L. R.: Blind Channel Equalization, IEEE Potentials, 18(4), 9–12, 1999. Campisi, P., Carli, M., Giunta, G., and Neri, A.: Blind Quality Assessment System for Multimedia Communications Using Tracing Watermarking, IEEE Transactions on Signal Processing [see also IEEE Transactions on Acoustics, Speech, and Signal Processing], 51(4), 996–1002, 2003. Ding, Y. W., Lin, Z., and Wang, L.: A Multipurpose PublicKey Cryptosystem Based Image Watermarking, 4th International Conference on Wireless Communications, Networking and Mobile Computing, WiCOM ’08., 2008.

www.adv-radio-sci.net/9/179/2011/

185

Benedetto, F., Giunta, G., and Neri, A.: End-to-End QoS Provision and Control in Wireless Communication Systems by Means of Digital Watermarking Signal Processing, 3rd International Symposium on Wireless Communication Systems, ISWCS ’06. 655– 659, 2006. Maity, S. P., Kundu, M. K., and Maity, S.: An Efficient Digital Watermarking Scheme for Dynamic Estimation of Wireless Channel Condition, International Conference on Computing: Theory and Applications. ICCTA ’07, 671–675, 2007. Neto, M. U., Gomes, L. de C. T., Romano, J. M. T., and Bonnet, M.: Adaptive Equalization based on Watermarking, 2006 International Telecommunications Symposium, 743–748, 2006. Zarzoso, V., and Comon, P.: Blind and Semi-Blind Equalization Based on the Constant Power Criterion, IEEE Transactions on Signal Processing, 53(11), 4363–4375, 2005. Sun, L., and Zhao, C.: Optimizing Blind Equalization Intelligent Algorithm for Wireless Communication Systems, 3rd International Conference on Natural Computation, ICNC 2007, 3, 146– 149, 2007. Dabeer, O., and Masry, E.: Convergence Analysis of the Constant Modulus Algorithm, IEEE Transactions on Information Theory, 49(6), 1447–1464, 2003. Podilchuk, C.I., and Delp, E.J.: Digital Watermarking: Algorithms and Applications, IEEE Signal Processing Magazine, 18(4), 3346, 2001. Samee, M. K., and G¨otze, J.: Increased Robustness and Security of Digital Watermarking Using DS-CDMA, Proceedings of the 7th IEEE International Symposium on Signal Processing and Information Technology, 189–193, 2007. Samee, M. K., G¨otze, J., Ruan, S. J., and Pai, Y. T.: Digital Watermarking: Spreading Code versus Channel Coding, Proceedings of the 3rd IEEE International Symposium on Communications, Control and Signal Processing, Malta, 1409–1412, 2008.

Adv. Radio Sci., 9, 179–185, 2011