Research On DSSS/OFDM Communication Techniques In Underwater Acoustic Network

Recently,the underwater acoustic network has been extensively investigated for o-ceanographic data collection.To achieve this goal,this network is envisaged to provide exchange of data between nodes,e.g.the main node sent the data require commands to the sensor node,and then the sensor node sent back the telemetry information.It need-s different communication requirements.For the commands transmission application,low rate and robust communication are sufficient;however high data rates are required for the telemetry data transmission application.Obviously,any modulation technique can't effectively satisfy the two kinds of communication requirements.Hence,the Direct-Sequence Spread-Spectrum(DSSS)is adopted for the low rate robust communi-cation and the Orthogonal Frequency Division Multiplex(OFDM)modulation is used to the high data rate communication.In DSSS modulation,a narrowband waveform is spread to a large bandwidth before transmission,which is one popular modulation scheme for low-rate robust communication;in contrast,the recently popular OFDM scheme is best suit for high data rate communications at high SNR scenarios.We in-vestigate the effective receiver design for underwater acoustic DSSS system and OFDM system,respectively,in order to overcome the negative effects of underwater acoustic channels,such as large propagation delay,fast time-varying and rich interference.The system input-output relationship is written based on parameterized path-delay channel model.We present the algorithms and the validations based on simulation and experi-mental data sets for the proposed receivers.The main contributions are as follows:·The novel Per-survivor processing based receiver design for DSSS communica-tion with a small spread.We firstly investigate a receiver for underwater DSSS transmissions which combines a PSP structure with sparse channel estimation.The PSP receiver maintains an excellent performance even for fast time-varying underwater acoustic channels.Then,the interest of this paper is on DSSS trans-missions with a small spreading factor in underwater acoustic communications.Such a system is desirable as data rate can be improved and transmission delay is reduced for DSSS when operating at low SNR level.As the decrease of the spreading factor,the existing PSP receiver has exponential complexity increase.As such,we propose a receiver with a set of parallel branches,which overcomes a major limitation of the existing PSP based receiver when the spreading factor is small.Note that all exiting publications on DSSS have used a spreading factor of several tens or several hundreds,rather than a single digit as in this work.Then,the channel coding module is applied in the DSSS system,which is not consid-ered in the existing PSP based receiver.Joint channel estimation and interference mitigation are performed solely on the parallel branches.The bit log-likelihood ratios from different branches are combined for channel decoding.Finally,sim-ulations and data sets collected from one field experiment held off the cost of Martha's Vineyard are used to validate the superior performance of the proposed receiver over the conventional RAKE receiver which is effective only when the spreading factor is large.·The novel receiver designs with impulsive noise mitigation in underwater acous-tic OFDM systems.First,we develop an effective non-iterative receiver design for joint Doppler shift and impulsive noise estimation based on a least-squares(LS)formulation with low complexity.Then,with available channel estimates and tentative symbol decisions,we further develop an iterative receiver.Finally,the two-component Gaussian Mixture(GM)model is used to fit the noise signals recorded from a real field.We have obtained useful experimental data sets where the received OFDM blocks contain impulsive noises.Performance results based on the data sets collected from one field experiment at the Taiwan strait,and also simulation with real impulsive noise characteristics,validate that the proposed receivers are very effective.In particular,the iterative receiver can approach the benchmark performance where the impulsive noise is absent.