Modeling and Simulation of P-Aloha, CSMA/CA and MACAW Protocols for Underwater Acoustic Channel

DAO Van-phuong, ZUO Jia-kuo (左加阔), ZHAO Li (赵 力), NGUYEN Dang-nam, ZOU Cai-rong (邹采荣), SUN Jian (孙 剑)

1 Key Laboratory of Underwater Acoustic Signal Processing of Ministry of Education, Southeast University, Nanjing 210003, China2 School of Information Engineering, Huangshan College, Huangshan 245021, China3 Petroleum Department, Petrovietnam University, Ba Ria 74000, Vietnam4 Ho Chi Minh City Industry and Trade College, Ho Chi Minh 760000, Vietnam

Modeling and Simulation of P-Aloha, CSMA/CA and MACAW Protocols for Underwater Acoustic Channel

DAO Van-phuong1,4*, ZUO Jia-kuo (左加阔)1, ZHAO Li (赵 力)1, NGUYEN Dang-nam3, ZOU Cai-rong (邹采荣)1, SUN Jian (孙 剑)2

1KeyLaboratoryofUnderwaterAcousticSignalProcessingofMinistryofEducation,SoutheastUniversity,Nanjing210003,China2SchoolofInformationEngineering,HuangshanCollege,Huangshan245021,China3PetroleumDepartment,PetrovietnamUniversity,BaRia74000,Vietnam4HoChiMinhCityIndustryandTradeCollege,HoChiMinh760000,Vietnam

Medium access control (MAC) protocol of underwater acoustic communication network is a key technology for underwater acoustic networks (UANs). Most of the MAC protocols for wireless terrestrial communication networks have been designed with negligible propagation delay. If it is deployed directly in an underwater environment, the UANs will perform inefficiently. In this paper, the characteristics of underwater acoustic channel are modeled and simulated by using the OPNET simulation tool, which are the speed of sound, propagation loss, and four sources for ambient noise: the turbulence, shipping, wind driven waves and thermal noise. The performance of pure Aloha (P-Aloha), carrier sense multiple access with collision avoidance (CSMA/CA) and multiple access collision avoidance for wireless local area network (MACAW) protocols in underwater acoustic channel environment are evaluated. The different performance of protocols in underwater environment is compared in the simulation.

underwateracousticchannel;OPNETsimulation;mediumaccesscontrol(MAC)protocol;pureAloha(P-Aloha)protocol;carriersensemultipleaccesswithcollisionavoidance(CSMA/CA)protocol;multipleaccesscollisionavoidanceforwirelesslocalareanetwork(MACAW)protocol

Introduction

The need for underwater communications exists in applications, such as military, oil industry, environment pollution, collection of scientific data, and discovery of new resources. Underwater communication networks can be established by using acoustic waves. Unlike terrestrial wireless networks that use radio waves for communication, underwater acoustic networks (UANs) utilize acoustic waves, which present many difficulties for both the physical and the data-link layers. One major disadvantage is that acoustic waves travel in underwater environment is very low (about 1500 m/s), which is five orders of magnitude slower than radio waves, and this leads to high propagation delay that reduces the throughput of UAN. The wave propagation in an underwater acoustic channel mainly gets affected by channel variations (temperature, salinity of water, pH of water, depth of water column or pressure, surface and bottom roughness), multipath propagation, and Doppler shift[1-7].

The practical design and deployment of UANs face some special challenges, one of which is the accurate underwater acoustic channel model for simulation.In Ref. [8], Jingetal. presented a shallow water acoustic channel model based on the actual acoustic propagation characteristics with path attenuation, ambient noise, multiple paths, and Doppler effects. In Ref. [9], the paper analyzed at-sea data collected in shallow water under various conditions to illustrate how the ocean environments could affect the signal properties. Channel properties is studied including amplitude and phase variations, and temporal coherence of individual paths as well as the temporal and spatial coherence of multipath at different time scales. In Ref. [10], the OPNET-based modeling of the underwater acoustic channel of the UANs was presented. In this work, Fanetal. mainly researched on the performance of the application of pure Aloha (P-Aloha) in UAN. In Ref. [11], the implementation of a channel model simulation using OPNET was presented, with respect to the Thorp model. In Ref. [12], Kingetal. presented an improved channel model based on NS-2, the development of an improved channel model based on the BELLHOP beam tracing program. In Ref. [13], Panetal. used the Thorp and Wenz models to simulate underwater acoustic channel. In Ref. [14], Bouzoualeghetal. researched on the modeling and simulation of the physical layer characteristics and communication protocol unit of the UAN communication systems with the state-flow and the SIMULINK simulation tools.

The researches above have focused on analyzing the characteristics of underwater environment[9], the performance of the application of an individual medium access control (MAC) protocol[10], channel model simulation, modeling and simulation of the physical layer characteristics[12-14]. Modeling underwater acoustic channel and simulation MAC protocols to estimate the performance of MAC protocols is very importance, which determinates the performance of MAC protocols in underwater acoustic channel. In this paper, the characteristics of underwater acoustic channel are modeled and simulated by using the OPNET simulation tool. The P-Aloha, carrier sense multiple access with collision avoidance (CSMA/CA) and multiple access collision avoidance for wireless local area network (MACAW) MAC protocols are simulated by using underwater acoustic channel, and the different performance of the protocols in underwater environment is compared in the simulation results.

The remainder of this paper is organized as follows. Section 1 describes the UAN model including of the propagation delay model, the propagation loss model, and ambient noise models that use to simulate UAN channel. Section 2 describes the P-Aloha, CSMA/CA and MACAW protocols. Section 3 is the simulation of the protocols. Section 4 is the results of simulation, and finally, and section 5 is the conclusions.

1 Modeling of the UAN Channel

In the OPNET, the wireless functionality of OPNET provides a special 14-stage transceiver pipeline used by the simulation Kernel to evaluate the characteristics of wireless communication, with the first six pipeline stages implemented in the transmitter and the rest eight in the receiver[11]. From Fig.1 we can see that several of the pipeline stages. For the UAN channel model which is used in this paper, the underwater acoustic channel is simulated with the characteristics of channel as follows: the Medwin’s speed equation for the speed of sound (stage 5); Fisher and Simons model for propagation loss (stage 7); and four sources for ambient noise are the turbulence, shipping, wind driven waves and thermal noise (stage 9).

Fig.1 Transceiver pipeline execution sequence for one transmission

1.1 Propagation delay (stage 5)

The speed of a wave propagating through a medium is not a constant. This is especially true for the non-homogeneous medium, the ocean. The speed of sound through water has been found to be mainly a function of three factors. They are temperature, pressure or depth and salinity. A simplified formula for the speed in m/s was given by Medwin in Ref. [15].

c =1449.2+4.6-0.055T2+0.00029T2+

(1.34-0.01T)(S-35)+0.016z.

(1)

whereTis temperature (℃),Sis the salinity in parts per thousand (‰),zis the depth (m), andcis the sound velocity (m/s).

1.2 Propagation loss (stage 7)

Propagation loss is composed majorly of three aspects, namely, geometrical spreading, attenuation and the anomaly of propagation. The attenuation, in dB, that occurs over a transmission rangelfor a signal frequencyfcan be obtained from[16-17]:

10log A(l, f)=k·10log l+l·10log α,

(2)

whereαis the absorption coefficient in dB/km, which can be obtained from Fisher & Simmons model, andkrepresents the geometrical spreading factor,k= 2 for spherical,k= 1.5 for practical andk= 1 for cylindrical.

The signal energy is one of the important factors that influence signal-to-noise ratio of receiver losses. The absorption loss of sound energy is the main part of the attenuation loss, and the absorption is usually seawater medium absorption and interface medium absorption. The Fisher & Simmons model is one of the most commonly used and referenced models[18-19].

(3)

wherefis the acoustic wave frequency. The coefficientsA1,A2,A3,P1,P2,P3,f1andf2in Eq. (3), can be obtained from Refs. [18-19].

1.3 Ambient noise model (stage 9)

The ambient noise in underwater can be described as Gaussion and having a continuous power spectral density. The four most prominent sources for ambient noise are the wind driven waves, shipping, turbulence, and thermal noises. The power spectral density for each of these is given by the formulae as follows[20].

10log Nt(f)=17-30log f,

(4)

10log Ns(f) =40+20(s-0.5)+26log f-

60log(f+0.03),

(5)

10log Nw(f) =50+7.5w1/2+20log f-

40log(f+0.4),

(6)

10log Nth(f)=-15+20log f.

(7)

The overall noise power spectral density may be obtained from:

N(f)=Nt(f)+Ns(f)+Nw(f)+Nth(f),

(8)

whereNtis the turbulence noise,Nsis the shipping noise (sis the shipping factor which lies between 0 and 1),Nwis wind driven wave noise (wis the wind speed in m/s), andNthis the thermal noise.

2 Protocols Description

2.1 P-Aloha protocol

P-Aloha protocol was proposed by Norman Abramson. In P-Aloha, all stations will be in idle state in usual time. When a station (sender) wants to send its data, it transmits data into channel, and enters the state of waiting for an acknowledgement (ACK) packet which is sent by the receiver to confirm the successful receipt of the data packet. At the receiver, when the proper receiver receives the data packet, it will send an ACK packet back to sender to acknowledge the receipt of the data packet. After a period of time, its timer is overtime and still hasn’t get the ACK packet, the sender will retransmit the data packet after a period of random back off time, and enters the state of waiting for ACK packet again. Due to the using of random time back off time, the probability of collision is reduced in the next transmission. If the retransmission number of the packet is up to the threshold, the transmission will be aborted[10, 21]. Figure 2 shows the P-Aloha protocol process model.

Fig.2 P-Aloha protocol process model

2.2 CSMA and CSMA/CA protocols

In the CSMA protocol, a station verifies the absence of other traffics before transmitting on a shared transmission medium. The sender tries to detect the presence of a carrier wave from another station before attempting to transmit. If a carrier is sensed, the station waits for the transmission in progress to finish before initiating its own transmission[22].

The CSMA/CA protocol is used to improve the performance of CSMA/CA by attempting to be less on the channel. If the channel is sensed busy before transmission then the transmission is deferred for a random interval time. This reduces the probability of collisions on the channel[23]. Figure 3 shows the CSMA/CA protocol process model.

Fig.3 CSMA/CA protocol process model

2.3 MACAW protocol

MACAW protocol was proposed by Vaduvur Bharghavan. This protocol uses three types of short, fixed-size signal packets. When station A wishes to transmit to station B, it sends a request-to-send (RTS) packet to B; this RTS packet contains the length of the proposed data transmission. If station B hears RTS packet, and it is not currently deferring, it immediately replies with a clear-to-send (CTS) packet; this CTS also contains the length of the proposed data transmission. Upon receiving the CTS, station A immediately sends its data. When station B receives a data packet from station A, it immediately sends an ACK packet to station A to notice that the data packet is received successfully[24-25]. Figure 4 shows the MACAW protocol process model.

Fig.4 MACAW protocol process model

3 Simulation of the Protocols

The simulations will be implemented for three MAC protocols that is P-Aloha, CSMA/CA and MACAW protocols. The network is a six-node static network and consists of one receiver node and five sender nodes. The receiver acts as the access point (AP) and receives data from the nodes 1-5. They are distributed in a region of 10 km width and 10 km length at 1000 m depth. The receiver node is in the surface of seawater. Each node is assigned an ID number as an address of its own while in intercommunication. Figure 5 shows the topology of the simulation network.

Fig.5 UAN topology

Generally, the simulation parameters are set in Table 1.

Table 1 Simulation parameters

ParametersValueTemperature/℃20Salinityofwater/‰35Speedofwind/(m·s-1)10Shippingdensity0.6Bandwidth/kHz10Fundamentalfrequency/kHz10

(Table 1 continued)

The packet generator generates 1000 bit packets that arrive at mean rate of 1 packet/s with an exponential inter-arrival time.

4 Simulation Results

In UAN, the throughput, average of end-to-end delay, average of channel utilization, collision status, and packet loss ratio are the important performance indexes. The following shows the simulation results under the multi-sequence simulation mechanism using OPNET tool.

In the P-Aloha protocol, whenever a node has a packet to send, it transmits that packet immediately, regardless of whether packet collisions are caused in its intended receiver. Due to the arbitrary transmissions of transmitter, the collision status of P-Aloha protocol at the receiver is higher than the CSMA/CA and MACAW protocols, as shown in Fig.6. The CSMA/CA protocol uses the carrier sense mechanism before transmitting data packet it makes the collisions at receiver to be lower than P-Aloha protocol. But due to the propagation delay in underwater environment, the collision status of CSMA/CA protocol is higher than MACAW protocol, which uses RTS, CTS, DATA, and ACK packets exchange, as shown in Fig.6.

Fig.6 Averages of collision status of P-Aloha, CSMA/CA and MACAW protocols in UAN

Figure 7 reflects the throughput performance comparison of three protocols. From Fig.7 it can be seen that the throughputs of these three protocols are quite low because the usable bandwidth of an underwater acoustic channel is limited, and the wave propagation in an underwater acoustic channel is mainly affected by channel variations, multi-path propagation, Doppler shift which keep lots of hurdles for achieving high data rates. The throughputs of the P-Aloha and CSMA/CA protocols are higher than the MACAW protocol due to the MACAW protocol uses RTS-CTS-DATA-ACK handshaking mechanism before transmitting data. The collision status of the MACAW protocol is also lower than the P-Aloha and CSMA/CA protocols, as shown in Figs.6 and 7.

Fig.7 Throughputs of P-Aloha, CSMA/CA and MACAW protocols in UAN

Figure 8 shows that the average of channel utilization of three protocols is quite low. The average of channel utilization of the P-Aloha and CSMA/CA protocols is around 54% and the MACAW protocol is about 34. The average of channel utilization of three protocols is low because the propagation delay of underwater acoustic channel is high while the available bandwidth of an underwater acoustic channel is limited. The average of channel utilization of MACAW protocol is much lower than the P-Aloha and CSMA/CA protocols because of the handshaking mechanism of MACAW protocol before transmitting data, which uses the RTS, CTS, DATA, and ACK packets exchange. The queue delay increases while the sound velocity in seawater is low, hence the end-to-end delay of the network increases. For MACAW protocol, the handshaking mechanism makes the end-to-end delay of this protocol to be higher than the P-Aloha and CSMA/CA protocols as shown in Fig.9.

Fig.8 Averages of channel utilization of P-Aloha, CSMA/CA and MACAW protocols in UAN

Fig.9 Averages of end-to-end delay of P-Aloha, CSMA/CA and MACAW protocols in UAN

As shown in Fig.10, the bit error rate is around 0.33, and it is high because the wave propagation in an underwater acoustic channel is mainly affected by channel variations, multi-path propagation, Doppler, temperature of water, shipping density, salinity, pH of water, depth of water column, surface and bottom roughness. Four basic noise sources as turbulence, shipping, wind driven waves, and thermal noise are the mainly noise sources which cause bit error rate in underwater acoustic channel.

Fig.10 Averages of bit error rate of P-Aloha, CSMA/CA and MACAW protocols in UAN

5 Conclusions

In this paper,an underwater acoustic channel model is designed using OPNET software according to the characteristics of underwater acoustic channel and implemented the simulation on the P-Aloha, CSMA/CA and MACAW protocols. The different performance of the protocols in underwater environment is compared in the simulation results. From the above simulation results, using MACAW and CSMA/CA protocols for UANs is not suitable. Due to these protocols use handshaking and carrier sense mechanisms before transmitting data packets, the collision status of MACAW and CSMA/CA protocols is low, and P-Aloha protocol is high. In MAC P-Aloha protocol, using retransmission mechanism when a node does not receive an ACK packet to response to the previous data packet. Hence the channel utilization and throughput of the P-Aloha protocol is much higher than CSMA/CA and MACAW protocols. These lead to end-to-end delay of P-Aloha to be lower than CSMA/CA and MACAW protocols. P-Aloha protocol is suitable to apply in underwater acoustic channel.

[1] Francois R E, Garrison G R. Sound Absorption Based on Ocean Measurements. Part I: Pure Water and Magnesium Sulfate Contributions [J].TheJournaloftheAcousticalSocietyofAmerica, 1982, 72(3): 896-907.

[2] Jensen F B, Kuperman W A, Porter M B,etal. Computational Ocean Acoustics [M]. New York: AIP Press, 1994.

[3] Urick R. Principles of Underwater Sound [M]. 3rd ed. New York: McGraw-Hill, 1983.

[4] Marsh H W, Schulkin M. Colossus II Shallow-Water Acoustic Propagation Studies [R]. USL Report No. 550, 1962.

[5] Francois R E, Garrison G R. Sound Absorption Based on Ocean Measurements. Part II: Boric Acid Contribution and Equation for Total Absorption [J].TheJournaloftheAcousticalSocietyofAmerica, 1982, 72(6): 1879-1890.

[6] Lin W, Chen E, Yuan F. An MACA-Based MAC Protocol for Underwater Acoustic Sensor Networks [J].JournalofCommunications, 2011, 6(2): 179-184.

[7] Han S, Noh Y, Uichin L,etal. M-FAMA: a Multi-session MAC Protocol for Reliable Underwater Acoustic Treams [C]. Proceedings IEEE INFOCOM, Turin, Italy, 2013: 665-673.

[8] Jing N, Bi W H, Yue Q. Attack Simulation Model and Channel Statistics in Underwater Acoustic Sensor Networks [J].TsinghuaScience&Technology, 2011, 16(6): 611-621.

[9] Yang T C. Properties of Underwater Acoustic Communication Channels in Shallow Water [J].JournaloftheAcousticalSocietyofAmerica, 2012, 131(1): 129-145.

[10] Fan G H, Wang D Q, Chen H B,etal. Simulation Research on P-Aloha Protocol in Underwater Acoustic Network Based on OPNET [C]. International Conference on Communications, Circuits and Systems, Xiamen, China, 2008: 153-156.

[11] Wang C, Fang Y J. Channel Model Simulation for Underwater Acoustic Sensor Networks Using OPNET [C]. The 12th IEEE International Conference on Communication Technology (ICCT), Nanjing, China, 2010: 141-144.

[12] King P, Venkatesan R, Li C. An Improved Communications Model for Underwater Sensor Networks [C]. IEEE Global Telecommunications Conference, New Orleans, LO, USA, 2008: 32-37.

[13] Pan C S, Jia L C, Cai R Y,etal. Modeling and Simulation of Channel for Underwater Communication Network [J].InternationalJournalofInnovativeComputing,InformationandControl, 2012, 8 (3B): 2149-2156.

[14] Bouzoualegh A, Val T, Campo E,etal. Modelling and Simulation of Underwater Acoustics Communication Based on Stateflow and Simulink Models [C]. The 3rd International Conference: Sciences of Electronic, Technologies of Information and Telecommunications, Tunisia, 2005: 1-7.

[15] Medwin H, Clay C S. Fundamentals of Acoustical Oceanography [M]. San Diego: Academic Press, 1998.

[16] Bai J, Liang Q, Yu H B,etal. Research on the Channel Simulation of Underwater Acoustic Networks [J].JournalofChineseComputerSystems, 2008, 29(1): 185-188. (in Chinese)

[17] Urban H G. Handbook of Underwater Acoustic Engineering [M]. Atlas Bremen: STN Atlas Elektronik GmbH, 2002.

[18] Fisher F, Simmons V. Sound Absorption in Sea Water [J].TheJournaloftheAcousticalSocietyofAmerica, 1977, 62(3): 558-564.

[19] Urick R J. Principles of Underwater Sound [M]. 3rd ed. Los Altos, California: Peninsula Publishing, 1983.

[20] Coates R. Underwater Acoustic Systems [M]. New York: John & Wiley, 1989.

[21] Abramson N. The ALOHA System-Another Alternative for Computer Communication [C]. Proceedings of the AFIPS Fall Joint Computer Conference, Las Vegas, USA, 1970: 281-285.

[22] Kleirock L, Tobagi F A. Packet Switching in Radio Channels: Part I — Carrier Sense Multiple-Access Modes and Their Throughput-Delay Characteristics [J].IEEETransactionsonCommunication, 1975, 28(12): 1400-1416.

[23] Karn P. MACA-a New Channel Access Method for Packet Radio [C]. ARRL/CRRL Amateur Radio 9th Computer Networking Conference, Ontario Canada, 1990: 1-5.

[24] Bharghavan V, Demers A, Shenker S,etal. MACAW: a Media Access Protocol for Wireless LAN’s [C]. Proceedings of the ACM SIGCOMM Conference, London, England, 1994: 212-255.

[25] Dao V P, Bao Y Q, Fang S L,etal. Slotted FAMA-CTS in MAC Protocol for Underwater Acoustic Networks [J].JournalofNanjingUniversity:NaturalSciences, 2013, 49(1): 78-85.

Foundation items: National Natural Science Foundations of China (Nos. 60872073, 6097501, and 51075068); the Doctoral Fund of Ministry of Education of China (No. 20110092130004); the Research Foundation and Education Bureau of Anhui Province of China (No. KJ2009B137)

TN912 Document code: A

1672-5220(2015)01-0035-06

Received date: 2013-10-09

* Correspondence should be addressed to DAO Van-phuong, E-mail: dao_van_phuong@yahoo.com