Encoded Total Focusing Method for Improving Data Acquisition Rate

C´esar Guti´errez-Fern´andez, Ana Jim´enez and Carlos Juli´an Mart´ın-Arguedas

Electronics Department, University of Alcal´a, Ctra. Madrid-Barcelona km. 33.600, Madrid 28805, Spain

Keywords:

Ultrasound Imaging, CDMA, Acquisition Rate.

Abstract:

Synthetic aperture imaging techniques are capable to obtain high quality images, fully focused in both trans-

mission and reception. However, these techniques require to perform so many emissions as elements in the

array to acquire RF data. This requirement decreases acquisition rate and can result in tissue motion artifacts

because of the phase misalignments between signals acquired in different emissions. Such inconvenience

claims for alternatives that reduce the total number of emissions needed to obtain the data. This work pro-

poses the use of Code Division Multiple Access (CDMA) techniques to attain this goal. By encoding the

ultrasound excitation signal emitted, through a pesudo-random Kasami code, several elements can emit simul-

taneously and the amount of data acquired in every emission increases. The encoded proposal attains a Kx

speed up in acquisition rate compared to conventional total focusing method, being K the number of Kasami

sequences available in the set.

1 INTRODUCTION

The use of ultrasound imaging systems based on ar-

rays has been widespread during the last decades,

standing out as a safer and cheaper alternative to other

imaging diagnosis techniques. Conventional Phased

Array (PA) imaging systems ﬁre all the elements of

the array in every excitacion, acquiring a line of the

image in every emission. The transmit beam must be

prefocused so images obtained blur out of the focus.

Synthetic Aperture Focusing Techniques (SAFT) rep-

resent an alternative to obtain fully focused images

(Jensen et al., 2006). In the basic SAFT approach the

transducer elements are activated sequentially acting

as both transmitter and receiver. One signal per emis-

sion is acquired and stored in memory. Once every

element has been activated and there are N signals

in memory, a beamforming process is applied to dy-

namically focus, both in emission and reception, ev-

ery pixel.

There are other SAFT approaches depending on

the number of active elements used in transmission

and reception. The SAFT method evaluated in this

work is the Total Focusing Method (TFM) (Lock-

wood and Foster, 1995), where a single element acts

as emitter and the echo data is recorded by the full

array. One of the drawbacks of SAFT techniques is

sensitivity to tissue motion during acquisition pro-

cess. Since Radio Frequency (RF) data is obtained

from multiple ﬁrings, motion artifacts may arise due

to phase misalignments between data from different

emissions. There are some solutions (Lokke Gam-

melmark and Jensen, 2003; Lokke Gammelmark,

2004) in the literature that propose motion compensa-

tion based on cross-correlation of reference signals to

ﬁnd the shift in position or phase of the tissue. These

proposals are capable to estimate motion tissue in 2D,

however their high computational cost is an inconve-

nience for real-time imaging sytems.

Another way to overcome motion artifacts is to re-

duce the acquisition time needed to form the image.

In this regard, encoding techniques can allows us to

achievethis goal by reducing the number of emissions

needed to acquire the data.

Encoding techniques and pulse compression have

been used in ultrasonic imaging mainly to enlarge

penetration depth and to improve SNR without in-

creasing the voltage excitation by using linear fre-

quency modulation (FM)(O’Donnel, 1992; Toosi and

Behnam, 2009). Some works have also studied

the application of encoding through binary codes.

(EMoo-Ho et al., 2002) propose Golay encoding to

obtain signal to ratio (SNR) improvement without re-

ducing frame rate. (Kim and Song, 2004) present

a linear array emitting Golay codes in combination

with chirp, achieving simultaneous multi-zone focus-

ing along several scan lines. (Romero-Laorden et al.,

2012) propose the use of Golay encoding in minimum

redundancy SAFT solutions to overcome their limited

SNR.

Gutiérrez Fernández C., Jiménez A. and Martín-Arguedas C..

Encoded Total Focusing Method for Improving Data Acquisition Rate.

DOI: 10.5220/0005255300950099

In Proceedings of the International Conference on Bioimaging (BIOIMAGING-2015), pages 95-99

ISBN: 978-989-758-072-7

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

Coded signals can be also used to differenti-

ate simultaneous emissions. These techniques are

known as Code Division Multiple Access (CDMA).

CDMA have been successfully used for indoor po-

sitioning and obstacle detection in robotics applica-

tions (De Marziani et al., 2012; Diego et al., 2012;

Klaus-Werner et al., 1998). In these works, a multi-

user scenario is considered, assigning a different code

to each user. Then simultaneous emissions and re-

ceptions from different users are possible, since each

code is used as an user identiﬁer.

Multiple access techniques have been already ap-

plied in ultrasonic imaging. (Shen, 1996) proposes to

emit a pseudo-orthogonal m-sequence in each direc-

tion, acquiring several lines at the same time thanks

to a bank of ﬁlters in reception stage. (Gran and

Jensen, 2008) propose the use of pseudo-random

codes in synthetic transmit aperture (STA) to separate

the echoes originating from two different transmitters

and improve SNR. Also (Diego et al., 2012) present

a design of an ultrasonic phased array with M emit-

ters and a single receiver based on encoded excita-

tion with loosely synchronous sequences for obstacle

detection, which scans the whole environment with a

single emission.

This work suggest a new method based on us-

ing CDMA techniques in TFM imaging to reduce

the emissions needed to acquire RF data. By encod-

ing the transmission with pseudo-orthogonal Kasami

codes several elements of the array can be excited si-

multaneously. Echoes coming back from the simul-

taneous emissions can be distinguished in reception

thanks to the suitable correlation properties of the

codes. Thereby the proposal allows to increase the

amount of data acquired by a single emission, acceler-

ating the acquisition process and making the imaging

system less sensible to motion artifacts.

2 KASAMI SEQUENCES

Kasami sequences are widespread in CDMA tech-

niques due to their suitable auto-correlation (AC) and

cross-correlation (CC) functions. The small set of

Kasami sequences used in this work is composed by

K = 2

X/2

sequences of length L = 2

− 1, where X

must be even. If the length of the sequences in-

creases, the number of sequences available also in-

creases. However, as it was exposed (Guti´errez-

Fern´andez et al., 2014), in ultrasound imaging appli-

cations code length is limited due to the blind-zone

problem. Therefore, 63 bits sequences are used in

this work, which can be used to scan distances begin-

ning from 5 cm. With this code length the Kasami set

is composed by K=8 pseudo-orthogonal sequences.

Generation of this Kasami sequences is based on the

algorithm proposed by (P´erez et al., 2009), which al-

lows to select those sequences with lower cross-talk

values in asynchronous environments.

3 TOTAL FOCUSING METHOD

In Total Focusing method a set of NxN signals is used

to obtain the highest image quality. A scheme of the

acquisition process is shown in Figure 1. In every

emission, a single element is ﬁred and all elements re-

ceive echo data. For a N-element array this process is

repeated N times, storing NxN recordings to compose

the image. Once stored in memory, these recordings

can be properly delayed to bring into focus all image

points.

Figure 1: Acquisition process in Total Focusing Method for

a 1-D N-element array. In every emission, a single element

is ﬁred and all elements receive echo data, acquiring N sig-

nals. This process is repeated for each array element, stor-

ing NxN recordings to compose the image.

TFM has the same T/R matrix and effective aper-

ture than conventional PA (Lockwood and Foster,

1995), but it is fully focused, so the highest image

quality is reached. Moreover, TFM reduces the elec-

tronics complexityin the emission stage by multiplex-

ing the same electronic in every emission. However,

as it has been aforementioned in section 1, TFM is

affected by motion artifacts due to there are phase

misalignments between signals acquired in different

emissions. Therefore, it is clear that a reduction in

the number of emissions needed to acquire RF data is

desirable to properly scan fast moving tissues.

BIOIMAGING2015-InternationalConferenceonBioimaging

4 APPLICATION OF CDMA

TECHNIQUES TO INCREASE

FRAME RATE

The acquisition scheme of the proposal is shown in

Figure 2, when a 1-D, N-element array and a set

of K Kasami sequences, C

,··· ,C

K−1

, are con-

sidered. In every emission there are K active ele-

ments, each one driven by a different Kasami code,

,··· ,C

K−1

. Orthogonality between codes as-

signed to each selected element makes possible that

all of them can emit simultaneously. The elements

activated in each emission are selected to maximize

the separation among simultaneous emissions, since

there is a separation of S-element between consecu-

tive active elements. In reception, the full array is

used to record N signals. Then, these signals are cor-

related with each one of the Kasami codes used in

emission, allowing us to distinguish echoes received

from the K simultaneous emissions. After correla-

tion K groups of N signals are obtained and stored

in memory. This process is repeated until the com-

plete set of NxN signals to compose the image is ob-

Figure 2: Acquisition process for one emission with the sys-

tem proposed, combining CDMA and TFM techniques. A

1-D, N-element array and a set of K Kasami sequences,

,··· ,C

K−1

, are considered. There are K active el-

ements in every emission, each one driven by a different

Kasami code, C

,... ,C

K−1

. The full array is used in

reception. KxN signals are acquired simultaneously in one

emission. A correlation process between the N signals and

the K codes emitted is carried out, obtaining KxN signals.

tained. Consequently, only S = N/K emissions must

be performed in this proposal, attaining a Kx speed up

in frame rate (imag/sec) respect to the classical TFM

technique, where N emissions are needed (Lockwood

and Foster, 1995).

5 SIMULATION RESULTS

Different simulations have been carried out to analyse

the advantages and drawbacks of the proposed tech-

nique in comparision with conventional TFM. An ar-

ray with N = 96 punctual elements, a pitch of half

wavelength and a frequency of 3 MHz are considered

for both cases. In all simulations punctual elements

with a ﬂat frequency response has been considered.

A Gaussian envelope pulse with 80% of relative

bandwidth has been considered as excitation for con-

ventional TFM images. In the encoded proposal, the

excitation signals are Kasami sequences with L = 63

bits, so the Kasami set is formed by K = 8 pseudo-

orthogonalsequences. The sequencesare BPSK mod-

ulated to adapt the codes to the working frequency of

the transducer, a single carrier period per bit has been

considered.

5.1 Contrast Resolution

Figure 3 showsthe images obtained with conventional

(a) and encoded (b) TFM proposals when a medium

with a punctual reﬂector located at r = 158 mm and

θ = 0

◦

is scanned. Each image has a resolution of

64x1200 pixels and it is fully focused, both in emis-

sion and reception, in each pixel. The encoded ap-

proach presents a background noise in the reﬂector

direction which reduces the image contrast in com-

parison with the conventional case. This background

noise is consequence of the interference inherent to

encoding techniques. The higher interference values

are 25 dB under the maximum response, these values

are mainly attributable to the AC interference.

5.2 Temporal Resolution

Conventional TFM needs to perform 96 emissions,

one per array element, to obtain the whole set of NxN

signals, while the enconded TFM requires only 12

emissions, therefore in the case shown in the ﬁgure

3 the proposal attains a 8x speed up in acquisition

rate respect to conventional TFM. The reduction of

the emissions needed to obtain RF data depends on

the number of pseudo-orthogonal codes availables in

the Kasami set. If the set is formed by K pseudo-

orthogonal codes, K elements can be ﬁred simultane-

EncodedTotalFocusingMethodforImprovingDataAcquisitionRate

Figure 3: Images obtained with conventional (a) and encoded (b) TFM proposals when a medium with a punctual reﬂector

located at r = 158 mm and θ = 0

◦

is scanned. Each image has a resolution of 64x1200 pixels. An array with N=96 elements

and pitch of half of wavelength is considered for both cases. Kasami sequences of L=63 bits are used to drive the elements in

the encoded approach.

ously in a single emission, so Kx speed up is reached.

This reduction in acquisition time can make the en-

coded TFM image system less sensible to motion ar-

tifacts and suitable to real-time applications.

6 CONCLUSIONS

This work has presented a proposal of synthetic aper-

ture technique TFM in conjunction with CDMA to

improve the acquisition rate. The central idea of the

proposal is that emission from multiple elements si-

multaneously can be achieved by driving them with

encoded signals, therefore, the RF data can be ac-

quired with fewer emissions. The excitation signals

used to drive the elements are a set of K pseudo-

orthogonal Kasami sequences.

Simulation results have shown that inherent inter-

ference of Kasami encoding adds a background noise,

reducing the image contrast respect to TFM reference

image. Further studies will be conduced to mini-

mize this interference by searching for other codes

with better correlation properties or applying post-

processing techniques.

The encoded TFM acquires the whole RF data

in only N/K emissions, whereas conventional TFM

requires N, one per array element. Consequently, a

Kx speed up in acquisition rate compared to conven-

tional method is attained. The improvement in acqui-

sition rates can be greater if the number of pseudo-

orthogonal codes available increases, since the num-

ber of elements that can emit simultaneously also in-

creases. The solution presented stands as a promising

alternative to make high quality TFM imaging sys-

tems less sensible to motion artifacts and capable to

reach acquisition time requirements in real-time ap-

plications.

ACKNOWLEDGEMENTS

This work has been granted by University of Al-

cal´a under cEYE project (CCG2013/EXP-043) and

the Spanish Ministry of Economy and Competitive-

ness under LORIS project (TIN2012-38080-C04-01)

REFERENCES

De Marziani, C., Ure˜na, J., Mazo, M., Hern´andez, A.,

Garc´ıa, J. J., P´erez, M. C., Aparicio, J., and Al-

coleas, R. (2012). Simultaneous round-trip time-of-

ﬂight measurements with encoded acoustic signals.

IEEE Sensors Journal, 12(10):2931–2940.

Diego, C., Martr´ın, C. J., Hern´andez, A., Jim´enez, A., and

Guti´errez, C. (2012). Improved ultrasonic phased ar-

ray based on encoded transmissions for obstacle de-

tection. IEEE Sensors Journal.

EMoo-Ho, B., Woo-Youl, L., Mok-Kun, J., and Sung-Jae,

K. (2002). Orthogonal golay code based ultrasonic

imaging without reducing frame rate. In IEEE Ultra-

sonics Symposium, volume 2, pages 1705–1708.

BIOIMAGING2015-InternationalConferenceonBioimaging

Gran, F. and Jensen, J. (2008). Spatial encoding using a

code division technique for fast ultrasound imaging.

IEEE transactions on ultrasonics, ferroelectris, and

frquency control, 55(1):12–23.

Guti´errez-Fern´andez, C., Jim´enez, A., Martr´ın, C. J.,

Hern´andez, A., and Ure˜na, J. (2014). A novel encoded

excitation scheme in a phased array for the improving

data acquisition rate. Sensors, 12:549–563.

Jensen, J. A., Nikolov, S. I., Gammelmark, K. L., and Ped-

ersen, M. H. (2006). Syntethic aperture ultrasound

imaging. Ultrasonics, 44:1–11.

Kim, B.-H. and Song, T.-K. (2004). Multibeam si-

multaneous transmit multizone (mb-stmz) focusing

method using modulated orthogonal codes for ultra-

sound imaging. pages 315–323.

Klaus-Werner, J., Berg, M., and M¨uller, M. (1998). Using

pseudo-random codes for mobile robot sonar sensing.

In 3rd IFAC Symposium on Intelligent Autonomus Ve-

hicles, Madrid, Spain, pages 25–27. IAV.

Lockwood, G. R. and Foster, F. S. (1995). Design of sparse

array imaging systems. In IEEE Ultrasonics Sympo-

sium, volume 2, pages 1237–1243.

Lokke Gammelmark, K. (2004). Improving the Image Qual-

ity of Synthetic Transmit Aperture Ultrasound Images.

PhD thesis, Technical University of Denmark.

Lokke Gammelmark, K. and Jensen, J. (2003). Duplex syn-

thetic aperture imaging with tissue motion compensa-

tion. In IEEE International Ultrasonics Symposium in

Honolulu, Hawaii, volume 2, pages 1569–1573 Vol.2.

O’Donnel, M. (1992). Coded excitation system for improv-

ing the penetration of real-time phased-array imaging

systems. IEEE T Ultrason Ferr.

P´erez, M. C., Ure˜na, J., Hern´andez, A., Jim´enez, A., Ru´ı,

D.,

Alvarez, F. J., and De Marziani, C. (2009). Perfor-

mance comparison of different codes in an ultrasonic-

positioning system using ds-cdma. pages 125–130.

Romero-Laorden, D., Martinez-Graullera, O., Martin-

Arguedas, C., and Parrilla-Romero, M. (2012). Ap-

plication of golay codes to improve snr in coarray

based synthetic aperture imaging systems. In Sensor

Array and Multichannel Signal Processing Workshop

(SAM), 2012 IEEE 7th, pages 325–328.

Shen, J. (1996). A new coded-excitation ultrasound imag-

ing system. IEEE Transanctions on Ultrasonics, Fer-

roelectrics, and Frecuency Control, 43(1):131–140.

Toosi, T. K. and Behnam, H. (2009). Combined pulse

compression and adaptive beamforming in coded ex-

citation ultrasound medical imaging. In International

Conference on Signal Processing Systems.

EncodedTotalFocusingMethodforImprovingDataAcquisitionRate