Communications – electrical: acoustic wave systems and devices – Underwater system – Telemetering
Reexamination Certificate
2002-05-14
2003-01-28
Lobo, Ian J. (Department: 3662)
Communications, electrical: acoustic wave systems and devices
Underwater system
Telemetering
C367S131000, C367S904000
Reexamination Certificate
active
06512720
ABSTRACT:
TECHNICAL FIELD
This invention relates to a method for phase coherent underwater acoustic communications. More particularly, the invention relates to a method for underwater acoustic communications using an improved Doppler compensation algorithm to extend the communication data packet length and reduce the number of receiver channels per fixed data and bit error rate.
BACKGROUND ART
Phase Coherent Underwater Acoustic Communications (ACOMMS)
The ocean presents an acoustic communication channel, which is band-limited and temporally variable. Propagation in the horizontal can be severely influenced by macro and micro multipath variability. Vertical propagation is often less severely impacted by the multipath.
Incoherent communication schemes, using for example frequency shift keying (FSK) algorithms, are used for line of sight propagation conditions in which multipath has minimal impact on the signals of interest. At long ranges, symbol rates for incoherent communications are limited by the multipath symbol interference. Additional processing (such as error encoding) is often required to remove the bit errors (due to symbol interference). The available frequency band is limited by frequency fading.
Coherent communication schemes use the available bandwidth more efficiently and provide higher data rates than the incoherent schemes for horizontal transmission of signals in a multipath environment. The state of the art systems use a (recursive) minimum least-mean-square (MLMS) approach for equalizing and updating the channel. The MLMS approach requires a certain minimum signal-to-noise ratio (SNR) at the receivers, of typically 10-15 dB. Maximum data rate and minimum bit error rate depend critically on the temporal properties of the channel impulse response function. The recursive least square (RLS) algorithm is computationally intensive and only a limited number (typically <4) of channels can be supported by prototype systems for real time communications.
An algorithm for phase coherent acoustic communications is described in U.S. Pat. Nos. 5,301,167 and 5,844,951, incorporated herein by reference. The latter patent extends the algorithm from a single receiver to multiple receivers; it uses jointly a phase locking loop and channel equalizer to adaptively correct for the channel temporal variation to minimize bit errors. The communication signals are transmitted by grouping symbols into packets. Each packet begins with a short pulse (e.g., a Barker code of 13 symbols of binary phases) used for symbol synchronization and an initial estimate of the multipath arrival structure. It is followed by a data packet beginning with a training data set with known symbols to estimate the carrier frequency (shift) and train the equalizer. The equalizer is updated by estimating the symbol errors using either the known symbol as in the training data or a decision on the received symbol. The number of tap coefficients is estimated from the impulse response deduced from the probe/trigger pulse. Carrier frequency s estimated from the training data. The data are fractionally sampled, typically 2 samples per symbol, and the most popular schemes for signal modulation are binary phase shift keying (BPSK) and/or quadrature phase shift keying (QPSK) signals. Channel impulse response and equalizer update requires a minimal input signal-to-noise (SNR) ratio for minimal bit errors. Multiple receivers using spatial diversity are often required for successful communications.
Underwater Acoustic Communication Channel
The underwater acoustic communication channel is different from the RF channel in three respects: (1) the long multipath delay due to sound refraction and long duration of reverberation from the ocean boundary; (2) the severe signal fading due to time-variable transmission loss; and (3) the high Doppler spread/shift, i.e., the variability and offset of receiver frequency and phase relative to the transmitter resulting from the media and/or platform motion.
In regard to the latter effect, underwater acoustic communications often involve sources or receivers from a moving platform. The source motion induces a shift in the carrier frequency, called Doppler shift. In practice, the frequency encounters not only a shift but also a frequency broadening; the latter referred to as Doppler spread. The sources of Doppler spread are twofold: sound scattering from moving ocean surface waves which has a broad spectrum and random small jitter of platform motion. The Doppler spread determines the signal coherence time assuming that the equalizer is able to update itself within the given coherence time. Because of these differences, the various techniques for radio frequencies (RF) communications cannot be applied directly to underwater acoustic communications.
Wireless radio communications are by line of sight with some multi-paths by reflection from nearby building and structure. Multi-path interference can usually be removed by antenna beamforming using an antenna on a horizontal plane. The array configuration can be designed with element spacing and configurations based on a plane wave model: the array aperture determines the width of the beam and element spacing determines the level of the side-lobes. Multipath delays in underwater acoustic channels, however, can last tens to hundreds of milliseconds, causing inter-symbol interference to extend over tens to hundreds of symbols depending on the carrier frequency and symbol rate. Inter-symbol interference in RF channels is orders of magnitude less and thus easier to deal with. Doppler shift of carrier frequency in underwater acoustic channels is several orders of magnitude larger than that of the RF channel since the sound speed is many orders lower than the speed of light. Hence, carrier frequency identification and symbol synchronization are critical for underwater acoustic communication systems. In addition, Doppler spread is non-negligible in the underwater communication channel as sound propagates through a random ocean medium and scatters from moving surfaces. Doppler spread and frequency coherence bandwidth accordingly limit the maximum data rate of underwater acoustic communications in an ocean channel.
Because Doppler shift can be relatively large in underwater acoustic communication channels, errors in the Doppler shift estimation (the carrier frequency identification) can cause large phase errors in phase modulated symbols and mis-synchronization of the symbol sequence, as the signal bandwidth (consequently the symbol duration) also changes proportionally with the carrier frequency shift. The estimation of carrier frequency (shift) in the prior art using the training data is limited by the duration of the training data, N multiplied by T, where N is the total number of training symbols and T is the symbol duration, providing a frequency resolution of the order I/NT. This frequency resolution is usually not precise enough, resulting in a symbol synchronization error over time. As a result, the communication packet length is limited (e.g., ≦5 sec at <5 kHz) beyond which the symbols are mis-aligned.
Ocean being a random medium with time varying surface boundary, acoustic signals propagating through the ocean encounter large random phase changes which must be removed or compensated for proper identification of phase modulated symbols. This was described above as the relatively large Doppler spread.
The channel equalizer can in principle track and compensate the random phase changes of each symbol. In reality, the equalizer requires tens of symbols to converge to a stable solution, as such it could not remove the random phase changes on a symbol-by-symbol basis. Thus a phase locking loop is used jointly with the channel equalizer to adaptively correct for the phase change and temporal variation of the channel impulse response. The coupling of the two operations produces a complicated inter-dependence between the phase compensation of the phase locked loop and the tap coefficients of the channel equalizer as they are being updated with the received
Karasek John J.
Legg Lawrence G.
Lobo Ian J.
The United States of America as represented by the Secretary of
LandOfFree
Underwater telemetry method using doppler compensation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Underwater telemetry method using doppler compensation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Underwater telemetry method using doppler compensation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3059152