A Multicarrier Digital Communication System for an Underwater Acoustic Environment

A Multicarrier Digital Communication System for an Underwater Acoustic Environment

Available online at www.sciencedirect.com ScienceDirect Procedia Technology 17 (2014) 625 – 631 Conference on Electronics, Telecommunications and Co...

552KB Sizes 3 Downloads 154 Views

Available online at www.sciencedirect.com

ScienceDirect Procedia Technology 17 (2014) 625 – 631

Conference on Electronics, Telecommunications and Computers – CETC 2013

A Multicarrier Digital Communication System for an Underwater Acoustic Environment Mário Lopesa, Renato Costaa, Paulo Marquesa,b a

Área Departamental de Engenharia Electrónica e Telecomunicações e de Computadores (ADEETC), Instituto Superior de Engenharia de Lisboa (ISEL), Rua Conselheiro Emídio Navarro 1, 1959-007 Lisbon, Portugal. b Instituto de Telecomunicações (IT), Instituto Superior Técnico, Avenida Rovisco Pais 1, 1049-001 Lisbon, Portugal.

Abstract Underwater acoustic applications are popular and vast, given the low attenuation of sound waves in water. However, these mediums are a challenge to efficient digital communication systems, due to their time and frequency dispersive nature. Considering the low bandwidth typically available in acoustic systems, high spectral efficiency is also required by the communication systems. In this article an Orthogonal Frequency Division Multiplexing (OFDM) modulation based solution is presented. Significant improvements to the bandwidth limitations and the time dispersion effects of the underwater channel are achieved through this technique. For the estimation and compensation of the Doppler spread a Least Mean Squares (LMS) adaptive algorithm is also considered, obtaining a spectral efficiency of 2.95 bit/second/Hertz. © 2014 The Authors. Authors. Published Published by by Elsevier Elsevier Ltd. Ltd. This is an open access article under the CC BY-NC-ND license © 2014 The (http://creativecommons.org/licenses/by-nc-nd/3.0/). Selection and peer-review under responsibility of ISEL – Instituto Superior de Engenharia de Lisboa. Peer-review under responsibility of ISEL – Instituto Superior de Engenharia de Lisboa, Lisbon, PORTUGAL. Keywords: Underwater acoustics; digital communications; OFDM; Doppler spread; LMS adaptive algorithm.

1. Introduction Most applications directed to underwater environments resort to acoustic signals, given their low attenuation in comparison with electromagnetic or optical signals. In search for higher data rates, digital communications in this medium are however limited by severe self-interference, caused by the time and frequency dispersive nature of these channels [1]. In shallow water, the slow speed of sound and considerable multipath phenomenon cause large delay spread, inducing Inter-Symbol Interference (ISI) in digital communications [1]. Significant ISI causes frequency-selective fading, reducing the coherence bandwidth of the underwater channel. Since most communications in this medium

2212-0173 © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). Peer-review under responsibility of ISEL – Instituto Superior de Engenharia de Lisboa, Lisbon, PORTUGAL. doi:10.1016/j.protcy.2014.10.185

626

Mário Lopes et al. / Procedia Technology 17 (2014) 625 – 631

have wideband characteristics, given the common bandwidth and carrier frequency limitation of acoustic systems, additional complexity in channel equalization is therefore required by the receiver. Relative motion induced by the movement of the transducers or changes in the underwater medium, may cause significant Doppler spread during communication [2]. Given the dynamic characteristic of the underwater medium, significant debate about the most adequate technique for compensation of this effect still occurs among the scientific community. In this context, the present article describes a digital communication in a shallow water acoustic medium, aiming for an improved performance over the limitations provided by these channels. 2. Experimental Setup For the desired tests in a shallow water environment, a pool tank was used with 9 meters length, 4 meters width and variable depth from 1.2 meters to 1.9 meters. The transmission and reception of the underwater acoustic signals are supported by a generic platform based on a Field-Programmable Gate Array (FPGA), which operates directly over the modulated signals. The signal generation and data processing are done by a Personal Computer (PC), communicating with the platform. As illustrated in Fig. 1, the electro-acoustic transducers were the only elements submerged, placed at 2 meters width, 0.6 meters depth and 9 meters of distance between them. The transducers used were the 200LM450 model from ProWave®, having a 200 kiloHertz central frequency and a 25 kiloHertz bandwidth.

Fig. 1: Experimental setup.

Copper wiring was used between the transducers and the platform. The largest cabled distance (of 29.5 meters) was covered by the transmitted signal, on behalf of reducing the received signal degradation, according to the Friis formula for noise factor [3]. On the experimental trials, the underwater medium had an average temperature of 20.3 degree Celsius, with a salinity of 0.5 parts per thousand and an acidity of 7.2 pH. Considering the homogeneity of this medium, according

Mário Lopes et al. / Procedia Technology 17 (2014) 625 – 631

the 1981 Mackenzie model, a constant speed of sound of 1483 meters per second is estimated [4]. No significant wind activity occurred during these tests. 3. Channel Characterization Given the shallow water characteristic of the used environment, it is known that the propagation of the transmitted acoustic waves is directly influenced by the geometry of the medium, since the wave reflections on the surfaces frontiers is the predominant effect. This aspect can be verified through the Power Delay Profile (PDP) depicted on Fig. 2, where the delay and amplitude of each multipath component is represented.

Fig. 2: Power Delay Profile of the underwater medium.

The primary component corresponds to the acoustic wave that has been transmitted directly between the transducers, having therefore the smallest path without any reflection. Due to this fact, this component has the highest amplitude of the received components, serving as the primary delay reference in Fig. 2. It can be inferred that the first and second component come from acoustic waves that propagate in a horizontal plane, reflecting with the lateral surfaces of the medium. These components are attenuated in relation to a third component, since it most likely corresponds to an acoustic wave that was reflected in the underwater surface, which typically is a good acoustic reflector [2]. A fourth component is almost simultaneous with the previous wave. Considering the approximate symmetry of the medium and reduced amplitude of this component, its origin is probably a reflection in the pool bottom. Although other components are visible in Fig. 2, 93.37% of the received signal energy is contained in the first 4 components, estimating therefore a delay spread of approximately 7 milliseconds, a value consistent with typical delay spread values, as presented in [2] and [5]. Regarding the frequency dispersive nature of the channel, it was concluded that the received signals did not suffer significant Doppler spread. This result seems consistent, since there was no intentional motion of the transducers and, considering the low wind activity during the trials, the agitation of the underwater medium was weak.

627

628

Mário Lopes et al. / Procedia Technology 17 (2014) 625 – 631

4. Digital Communication 4.1. Initial considerations The OFDM technique was the chosen modulation for the digital communication in the studied underwater acoustic channel, since it provides efficient techniques to address the dispersive nature of this medium in the time domain. For this effect, OFDM allows the increasing of the symbol duration without reducing the achieved data rate. Signal distortion caused by ISI can also be effectively mitigated through a guard time larger than the delay spread of the channel. OFDM symbol uses orthogonal subcarriers for parallel data transmission. In the presence of Doppler spreading, however, the orthogonality of these subcarriers can be compromised, resulting in Inter-Carrier Interference (ICI). Although the used medium had not a highly dispersive nature in this domain, a LMS adaptive algorithm for the estimation and compensation of this effect was considered, which is described in reference [6]. This algorithm is applied after the symbol synchronism, not requiring previous channel information since the Doppler spread estimation is based on the received pilot carriers. 4.2. Main results Tab. 1 summarizes the main parameters of the digital communication in the underwater medium. These results were obtained by the experimental setup described in Section II. Table 1. Main Parameters of the Digital Communication Bandwidth Limit frequencies Symbol duration Guard interval Subcarrier spacing Number of data subcarriers Data subcarriers modulation Number of pilot carriers Channel interpolation Synchronism Peak average power ratio Data rate

35 kHz 180 kHz - 215 kHz 524.288 ms 8.192 ms 1.907 Hz 13750 16-QAM 4587 Linear Cyclic Prefix 8.35 dB 103.288 kbit/s

Considering the bandwidth limitation of the acoustic transducers (25 kiloHertz), spectral efficiency was increased through the orthogonal characteristic of the OFDM subcarriers. Despite the wideband nature of the communication, channel equalization implies low complexity, assuming linear interpolators. Based on its satisfactory performance, the occupied bandwidth was extended to 35 kiloHertz. With a guard interval larger than the delay spread of the channel (7 milliseconds) and high symbol duration, significant improvement to ISI was achieved without major impact on the data rate. A cyclic prefix was used in the guard interval for symbol synchronism. With this technique it was ensured a continuous activity of the transducers, reducing signal distortion caused by the ringing of these components [7]. Peak Average Power Ratio (PAPR) represents a major disadvantage in OFDM systems. To avoid clipping distortion in the amplifiers, signal power is typically reduced, minimizing however the Signal-to-Noise Ratio (SNR) [8]. In this system, PAPR was effectively reduced to 8.35 decibel by a random phase codification with uniform distribution in the pilot carriers. As showed in Fig.3, in the data subcarriers a 16-Quadrature Amplitude Modulation (QAM) was achieved, resulting on a data rate of 103.288 kilobits/second [9] with a Bit Error Rate (BER) of 0.004091. Considering the transmitted signals were modulated in Single Side Band (SSB), a spectral efficiency of 2.95 bit/second/Hertz was obtained.

Mário Lopes et al. / Procedia Technology 17 (2014) 625 – 631

Fig. 3: From left to right, received 16-QAM constellation before and after channel equalization.

Considering the stationarity of the medium, binary decoding was not compromised without ICI compensation. However, the Doppler spread was estimated using the considered LMS algorithm presented in [6].

Fig. 4: Doppler spread estimation of the proposed algorithm (step size of 0.000005).

Fig. 4 presents the learning curve of the LMS algorithm. A Doppler spread estimation of 0.5683 Hertz was achieved, resulting in a BER of 0.004164. Considerable ICI would be expected, given the subcarrier spacing of 1.907 Hertz. This result however is not coherent with the previous channel characterization and BER performance, leading the authors to the conclusion that there is an error in the Doppler spread estimation. This divergent result was possibly induced by an inadequate channel model used in the algorithm, which accounts for a linear Doppler spreading. Since non-linear Doppler variations have been described in some underwater acoustic channels [10], this effect and other additional distortion induced by the channel could have been the source of error. Further tests are required to validate these deductions.

629

630

Mário Lopes et al. / Procedia Technology 17 (2014) 625 – 631

4.3. Results comparison An approach for develop efficient OFDM systems in underwater acoustic channels is presented in this article. The approach suggests a development of a communication system based on two sources: experimental characterization and scientific literature. A performance comparison between different communication systems in a shallow water environment is made in Tab. 2, based on achieved data rate and spectral efficiency. All presented results are referenced to the same BER order. Table 2. Result Comparison with other Communication Systems.

Spectral Efficiency [bit/s/Hz] Data rate [bit/s]

16QAM-OFDM [Presented System] 2.95 103288

FSK [11] 0.05 75

QPSK [12] 1 2200

16QAM-OFDM [9] 1 50000

Since single carrier modulations ([11], [12]) are severely challenged by ISI, the presented system obtains a better performance. This system also had a better performance in comparison with the other multicarrier system. This improvement can be explained by effective mitigation of ISI, considering the information obtained though experimental characterization. The use of SSB modulation and the adequate performance of channel interpolators were equally important to the presented results, not recurring to any error coding to minimize the obtained BER [9]. Considerations between the chosen Doppler spread estimators are inconclusive with the current tests. 5. Conclusions In this article a successful OFDM based digital communication was achieved in an underwater acoustic channel. OFDM modulation was chosen because it presents efficient mechanisms to address the temporal dispersive nature of the underwater channel. Additionally, it allows simplified channel equalization and high spectral efficiency, addressing the wideband nature and the severe bandwidth limitation of this channel. Since PAPR (Peak Average Power Ratio) represents a major limitation in OFDM systems, a random phase codification scheme in the pilot carriers was considered, reducing the power ratio to 8.35 decibel. In the presence of Doppler spreading the orthogonality of the subcarriers could be compromised. A LMS adaptive algorithm was considered for estimation and compensation of this effect. Since this algorithm didn't contemplate nonlinear Doppler variations and other signal distortions provided by the underwater channel, inaccurate results were obtained. Using a 16-QAM modulation for the OFDM data subcarriers, a final data rate of 103.288 kilobits/second was achieved in a 35 kiloHertz channel, obtaining a BER (Bit Error Rate) of 0.004091 and a spectral efficiency of 2.95 bit/second/Hertz. References [1] [2] [3] [4] [5] [6] [7] [8] [9]

L. Liu, S. Zhou, and J. Cui, “Prospects and Problems of Wireless Communication for Underwater Sensor Networks”, Wiley J. Wireless Communications & Mobile Computing, vol. 8, issue 8, pp. 8, October 2008. M. Stojanovic and J. Preisig, “Underwater Acoustic Communication Channels: Propagation Models and Statistical Characterization”, IEEE Communications Magn., vol. 47, issue 1, pp. 84–89, 2009. H. T. Friis, “Noise Figures of Radio Receivers”, Proceedings of the IRE, vol. 32, pp. 419-422, July 1944. K. V. Mackenzie, “Nine-term equation for the sound speed in the oceans”, J. Acoust. Soc. Am, pp. 807-812, 1981. B. Tomasi, G. Zappa, K. McCoy, P. Casari and M. Zorzi, “Experimental study of the Space-Time Properties of Acoustic Channels for Underwater Communications”, IEEE OCEANS - 2010 Sydney, pp. 1–9, 2010. F. Yang, K. H. Li and K. C. Teh, “Adaptive LMS-like algorithm for Carrier Frequency Offset Estimation in OFDM Systems”, Proceedings of the 8th International Symposium on Signal Processing and Its Applications, vol. 1, pp. 131–134, 2005. L. F. Yeung, G. Nian and Z. Choujun, “An Inter-Carrier Interference Reduction Scheme for OFDM Underwater Acoustic Communications”, 9th International Symposium on Communications and Information Technology, pp. 1173 - 1177, 2009. V. B. Malode and B.P. Patil, “PAPR Reduction Using Modified Selective Mapping Technique”, Int. J. of Advanced Networking and Applications, vol. 2, issue 2, pp. 626-630, 2010 B. Li, S. Zhou, J. Huang and P. Willett, “Scalable OFDM design for underwater acoustic communications", IEEE International Conference on Acoustics, Speech and Signal Processing, 2008.

Mário Lopes et al. / Procedia Technology 17 (2014) 625 – 631 [10] A. C. Singer, J. K. Nelson and S. S. Kozat, "Signal processing for underwater acoustic communications", IEEE Communications Magn., vol. 47, issue 1, pp. 90–96, 2009. [11] R. F. W. Coates, “A deep-ocean penetrator telemetry system”, IEEE J.Oceanic Eng., vol. 13, pp. 55–63, 1988 [12] S. M. Jarvis and N. A. Pendergrass, “Implementation of a multichannel decision feedback equalizer for shallow water acoustic telemetry using a stabilized fast transversal filters algorithm”, presented at the Oceans’95, San Diego, CA, 1995.

631