Underwater communications apparatus and method of making an improved estimate of doppler shift

Abstract

Doppler shift estimation of an underwater acoustic signal 204. The underwater acoustic signal is transmitted as a packet including a first waveform portion (142, Fig. 2), i.e. a preamble portion, and a second waveform portion 144, such as a physical layer control signal portion. Complex baseband data is derived from a received acoustic signal 204 and buffered 222. The first waveform portion is applied to a correlator bank 224 to identify the preamble portion and estimate a first Doppler shift. The first and second waveform portions are coarse Doppler compensated using the first Doppler shift estimate and the resulting compensated first waveform portion is correlated with a replica of the first waveform portion, thereby generating a preamble correlation output signal. The compensated second waveform portion is decoded by generating 228 a maximum likelihood correlation signal associated with the decoded compensated second waveform portion. A residual Doppler shift is calculated using the preamble correlation output signal and the maximum likelihood signal and a Doppler shift estimate is calculated 230 using the calculated residual Doppler shift and the first estimate of the Doppler shift.

Description

[0001] The present invention relates to a method of making an improved estimate of Doppler shift, the method being of the type that, for example, comprises receiving an acoustic communications signal in an underwater environment, the acoustic communications signal experiencing a Doppler shift. The present invention also relates to an underwater communications apparatus of the type that comprises, for example, an underwater acoustic transducer for receiving an acoustic communications signal that experiences a Doppler shift.

[0002] In underwater acoustic communications systems, acoustic signals experience so-called Doppler shift owing to various factors including intrinsic properties of the underwater environment and mechanical aspects, such as relative motion of acoustic transmitters and receivers, irrespective of the data structure to which the acoustic signal conforms.

[0003] When transmitting an acoustic signal for positioning purposes, for example for calculating a time of arrival of an acoustic signal at an underwater communications apparatus, it is necessary to take into account, and compensate for, the Doppler shift that the transmitted acoustic signal has experienced. Otherwise, the result of any subsequent calculation, for example a range calculation, will be distorted.

[0004] In recent years, it is also known to use underwater acoustic signals to communicate telemetry data in order to support communication of command-and-control data in a multi-user environment, for example as suggested in “Multiuser Navigation and Telemetry Performance Issues in Shallow Water Environments” (JJ Davies et al., Oceans 2009-Europe, IEEE, 11-14 May 2009, pages 1-10). A key difference between an acoustic signal used for underwater positioning as opposed to telemetry is that, for telemetry, the underwater acoustic signal generally comprises multiple waveforms rather than a single waveform, and so more information can be encoded within these waveforms. Underwater acoustic communications are no different to other communications and motional Doppler is present on each propagating path component according to each path’s specific geometry. Uncorrected, motional Doppler leads to time coherence loss in wideband coherent acoustic systems, reduced energy, and degraded detection and parameter estimation performance. For this reason, all coherent acoustic telemetry systems, rely on motional Doppler estimation and compensation as part of initial or ‘open loop’ detection-synchronisation.

[0005] It is also known to combine both telemetry and positioning objectives through use of an underwater communications system that employs a relatively short packet structure, which extends basic positioning functionality, to include support for a data payload. A spread spectrum communications technique is typically employed to propagate packets and, in one example, the packet structure comprises a first symbol, a second symbol and a payload. The first symbol serves a detection purpose, and the second symbol provides information concerning downstream physical layer processing so that a receiver can switch to appropriate telemetry de-modulation and channel decoding modes to recover information in the payload of the packet that follows the second symbol. The size of the payload part of the packet is typically of the order of a few bytes.

[0006] Such packet structures enable the communication of command-and-control data amongst multiple users, whilst also enabling positioning measurements to be made, for example a time of arrival of the packet. In order to account for Doppler shift, a replica of the first symbol in multiple frequency shifted forms is respectively applied to a bank of correlators along with complex baseband data of a received acoustic signal obtained following digital down conversion. The outputs of the bank of correlators can have peaks of differing amplitude, and these are analysed. For example, an output signal from a correlator having a highest peak amplitude, indicative of a highest level of correlation between a replica of the first waveform and the part of the complex baseband data corresponding to where the first symbol should be if the symbol is present, can be selected and the frequency shift associated with the replica responsible for the highest peak provides an estimate of the Doppler shift experienced by the received acoustic signal. In other examples, the output signals from a number of correlators of the bank of correlators can be selected based upon a closest degree of correlation and processed, for example by way of an interpolation technique, in order to obtain a “true” position of a highest peak and hence an improved estimate of Doppler shift.

[0007] However, whilst such techniques provide an adequate estimate of Doppler shift for many applications, and competent delivery of telemetry data, it is desirable for a positioning parameter, such as time of arrival of the received acoustic signal, to be as precise as possible. In this regard, precision of the estimate of Doppler shift directly impacts precision of estimation of time of arrival and phase angle of arrival. As such, it is desirable to improve precision of estimation of Doppler shift, and in turn precision of estimation of one or more positioning parameter.

[0008] According to a first aspect of the present invention, there is provided a method of making an improved estimate of Doppler shift by an underwater communications apparatus, the method comprising: receiving an acoustic communications signal and buffering complex baseband data derived therefrom, the complex baseband data representing a plurality of symbols organised as a first waveform portion and a second waveform portion conforming to a data structure definition; applying the first waveform portion of the complex baseband data to a first bank of correlators to identify a preamble waveform portion; calculating a first estimate of a Doppler shift in respect of the first predetermined waveform portion; coarse Doppler compensating the first and second waveform portions of the complex baseband data for the Doppler shift thereof using the first estimate of the Doppler shift; correlating the coarse Doppler-compensated first waveform portion of the complex baseband data with a replica of the first waveform portion, thereby generating a preamble correlation output signal; decoding the coarse Doppler-compensated second waveform portion of the complex baseband data, the decoding of the coarse Doppler-compensated second waveform portion comprising generating a maximum likelihood correlation output signal associated with the decoded coarse Doppler-compensated second waveform portion; calculating a residual Doppler shift using the preamble correlation output signal and the maximum likelihood correlation output signal; and calculating the improved estimate of the Doppler shift using the calculated residual Doppler shift and the first estimate of the Doppler shift.

[0009] The plurality of symbols may be further organised as the first waveform portion, the second waveform portion and a third waveform portion conforming to the data structure definition.

[0010] The method may further comprise: decoding the coarse Doppler-compensated second waveform portion of the complex baseband data to obtain a control code.

[0011] The method may further comprise: applying the coarse Doppler-compensated second waveform portion of the complex baseband data to a second bank of correlators in order to decode the coarse Doppler-compensated second waveform portion of the complex baseband data.

[0012] The second bank of correlators may be configured to correlate the coarse Doppler-compensated second waveform portion of the complex baseband data with a plurality of candidate replica waveform portions.

[0013] The method may further comprise: calculating the maximum likelihood correlation output signal from application of the coarse Doppler-compensated second waveform portion of the complex baseband data to the second bank of correlators.

[0014] The method may further comprise: the first bank of correlators generating a plurality of correlation output signals, each of the plurality of correlation output signals resulting from correlation of a different frequency-shifted version of a replica of the first waveform portion with the first waveform portion of the complex baseband H ofo lldld.

[0015] The method may further comprise: selecting a first correlation output signal from the plurality of correlation output signals and in respect of a first frequency-shifted version of the replica of the first waveform portion, the first correlation output signal being indicative of a highest degree of correlation with respect to other correlation output signals of the plurality of correlation output signals; selecting a second correlation output signal from the plurality of correlation output signals and in respect of a second frequency-shifted version of the replica of the first waveform portion, the second correlation output signal being indicative of a second highest degree of correlation with respect to the other correlation output signals of the plurality of correlation output signals; selecting a third correlation output signal from the plurality of correlation output signals and in respect of a third frequency-shifted version of the replica of the first waveform portion, the third correlation output signal being indicative of a third highest degree of correlation with respect to the other correlation output signals of the plurality of correlation output signals; and calculating the first estimate of the Doppler shift using the first, second and third correlation output signals selected.

[0016] The first, second and third correlation output signals may each comprise a respective first, second and third maximum value; and the calculating of the first estimate of the Doppler shift may further comprise: interpolating the first, second and third maximum values to estimate an actual frequency at which maximum correlation occurs.

[0017] The method may further comprise: calculating the first estimate of the Doppler shift using the estimated actual frequency at which the maximum correlation occurs.

[0018] The method may further comprise: fine Doppler compensating the third waveform portion of the complex baseband data using the improved estimate of the Doppler shift; and downconverting and decoding the third waveform portion of the fine Doppler-compensated complex baseband data in accordance with the control code.

[0019] The data structure definition may comply with an error correction technique; and the method may further comprise: decoding the third waveform portion of the fine Doppler-compensated complex baseband data corresponding to the third waveform portion using the error correction technique.

[0020] According to a second aspect of the present invention, there is provided a method of calculating an underwater navigation parameter using a received acoustic communications signal, the method comprising: calculating the improved estimate of the Doppler shift in respect of the received acoustic communication signal as set forth above in relation to the first aspect of the present invention; fine Doppler compensating at least the first and second waveform portions of the complex baseband data using the improved estimate of the Doppler shift to provide fine Doppler-compensated complex baseband data; and applying the fine Doppler-compensated complex baseband data to a bank of matching correlators, the bank of matching correlators correlating at least the first waveform portion of the fine Doppler-compensated complex baseband data with the first replica thereof and the second waveform portion of the fine Doppler-compensated complex baseband data with a second replica thereof in order generate a reprocessed correlation output signal; and calculating the underwater navigation parameter in respect of the received acoustic communications signal using the reprocessed correlation output signal.

[0021] The fine Doppler-compensated complex baseband data may include at least part of the third waveform portion of the fine Doppler-compensated complex baseband data compensated using the improved estimate of the Doppler shift; and the method may further comprise: the bank of matching correlators additionally correlating the at least part of the third waveform portion of the fine Doppler-compensated complex baseband data with a third replica thereof, thereby contributing to the generation of the reprocessed correlation output signal.

[0022] The downcoverted and decoded at least part of the third waveform portion of the fine Doppler-compensated complex baseband data may serve as the third replica of the fine Doppler-compensated baseband data.

[0023] The underwater navigation parameter may be a time of arrival. The underwater navigation parameter may be a phase angle of arrival.

[0024] The method may further comprise: calculating a plurality of signal quality indicator values in respect of a plurality of symbols decoded from the third waveform portion of the fine Doppler-compensated complex baseband data, respectively; and selecting a number of symbols from the plurality of symbols having respective signal quality indicator values conforming to a predetermined criterion to serve as the at least part of the third waveform portion of the fine Doppler-compensated complex baseband data.

[0025] According to a third aspect of the present invention, there is provided a method of calculating a bearing of arrival of an acoustic communications signal, the method comprising: receiving an acoustic communications signal at a plurality of acoustic transducers; and calculating a phase angle of arrival in respect of each of the plurality of acoustic transducers using the method of calculating the underwater navigation parameter as set forth above in relation to the second aspect of the present invention; and calculating a bearing of arrival using the plurality of phase angles of arrival.

[0026] The method may further comprise: transmitting an acoustic communications interrogation signal to another underwater communications apparatus; the another underwater communications apparatus may transmitting the acoustic communications signal in response to receipt of the acoustic communications interrogation signal; and recording a time of transmission of the acoustic communications interrogation signal.

[0027] The method may further comprise: calculating a round trip time in respect of transmission of the acoustic communications interrogation signal using a time of transmission of the acoustic communications interrogation signal, a time of arrival of the acoustic communications interrogation signal, a turnaround time of the another underwater communications apparatus, and the calculated time of arrival of the received acoustic communications signal; and estimating a time-of-flight in respect of the received acoustic communications signal.

[0028] According to a fourth aspect of the present invention, there is provided an underwater communications apparatus for making an improved estimate of Doppler shift, the apparatus comprising: an underwater acoustic transducer configured to receive an acoustic communications signal; a data store to buffer complex baseband data derived from the received acoustic communications signal, the complex baseband data representing a plurality of symbols organised as a first waveform portion and a second waveform portion conforming to a data structure definition; a signal processing resource comprising a first bank of correlators operably coupled to the underwater acoustic receiver; a Doppler estimator operably coupled to the first bank of correlators and configured to calculate a first estimate of a Doppler shift in respect of the first predetermined waveform portion; a Doppler compensator operably coupled to the data store and the Doppler estimator and configured to coarse Doppler compensate the first and second waveform portions of the complex baseband data for the Doppler shift thereof using the first estimate of the Doppler shift; a preamble correlator configured to correlate the coarse Doppler-compensated first waveform portion of the complex baseband data with a replica of the first waveform portion, thereby generating, when in use, a preamble correlation output signal; a decoder configured to generate a maximum likelihood correlation output signal associated with decoding the coarse Doppler-compensated second waveform portion of the complex baseband data; the Doppler estimator is configured to calculate a residual Doppler shift using the preamble correlation output signal and the maximum likelihood correlation output signal; and the Doppler estimator is configured to calculate the improved estimate of the Doppler shift using the calculated residual Doppler shift and the first estimate of the Doppler shift.

[0029] It is thus possible to provide an apparatus and method that exhibits an improved signal-to-noise ratio in respect of correlation processing (processing gain), and thereby providing estimates of time of arrival and phase angle of arrival with improved precision. In this regard, the ability of the apparatus and method to precisely estimate Doppler shift contributes to the improved processing gain achieved.

[0030] At least one embodiment of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 is a schematic diagram of an underwater environment comprising a variety of vessels, instruments and personnel possessing underwater communications devices configured to operate in an underwater communication system; Figure 2 is a schematic diagram of a structure of a packet used by the underwater communications system of Figure 1; Figure 3 is a schematic diagram of a simplified communications arrangement between a first underwater communications device and a second underwater communications device in the underwater environment of Figure 1; Figure 4 is a schematic diagram of a signal processing arrangement for generating digital baseband data from a received acoustic communications signal and generating a coarse estimate of a Doppler frequency shift, constituting an embodiment of the invention; Figure 5 is a schematic diagram of another signal processing arrangement configured to provide coarse Doppler compensation and subsequent waveform correlation, and constituting another embodiment of the invention; Figure 6 is a schematic diagram of a further signal processing arrangement configured to Doppler compensate and decode a payload and constituting yet another embodiment of the invention; Figure 7 is a schematic diagram of a signal re-processor constituting a further embodiment of the invention; Figure 8 is a schematic diagram of a bearing of arrival estimator constituting another embodiment of the invention; Figure 9 is a flow diagram of a communication sequence for measuring a navigation parameter constituting further embodiment of the invention; Figure 10 is a flow diagram of a first part of a method of making an improved Doppler estimate constituting yet a further embodiment of the invention; Figure 11 is a flow diagram of a second part of the method of Figure 10; Figure 12 is a flow diagram in part of a third part of the method of Figure 10, as well as calculation of a navigation parameter in accordance with yet another embodiment of the invention; and Figure 13 is a flow diagram employing phase angles of arrival calculated in Figure 11 in order to calculate a bearing of arrival.

[0031] Throughout the following description identical reference numerals will be used to identify like parts.

[0032] Referring to Figure 1, an underwater environment 100 comprises a water column 102 having differing properties that vary with location within the water column 102, for example temperature, pressure and salinity. These properties affect the propagation of acoustic sound waves. The underwater environment 100 is between a surface, for example a sea surface 104, and a bottom, for example a seabed 106. A number of vessels can be located at the sea surface 104 in communication with or capable of communication with vessels, divers and / or instruments below the sea surface, for example a first vessels 108, a second vessel 110 and a third vessel 112.

[0033] In this example, the first vessel 108 is in communication with a first Autonomous Underwater Vehicle (AUV) 114. The first AUV 114 is also in communication with another form of AUV, for example an eXtra-Large Unmanned Underwater Vehicle (XLUUV) 116, as well as a first seabed instrument, for example a first seabed pressure sensor 118. In this example, the first seabed pressure sensor 118 is in communication with a second seabed instrument, for example a second seabed pressure sensor 120, the second seabed pressure sensor 120 also being in communication with a communications buoy, for example a gateway buoy 122. The gateway buoy 122 can be employed to provide a communications link between underwater communications devices and surface-side communications devices, for example ground stations and aerial devices, or space-based devices, such as a satellite 124.

[0034] The second vessel 110 is also in communication with a number of divers 126, each of the divers 126 carrying a respective communications device. The divers 126 are therefore also capable of communicating with each other using their respective communications devices. Further seabed instruments, for example subsea nodes 128, are disposed on the seabed 106. The subsea nodes 128, for example, seismic seabed nodes, are capable of communicating with each other, as well as other devices, for example a second AUV 130.

[0035] The third vessel 112 is, in this example, in communication with another class of underwater vehicle, for example a submarine 132. In the present example, the underwater communications between communications devices carried by personnel and / or vessels and / or instruments are configured to operate in conformity with a technical specification of an underwater communications system that supports combined positioning and telemetry data functionality. To provide this functionality, the communications system transmits and receives data in accordance with a data structure definition, for example a packet structure. Referring to Figure 2, and treating the entire packet as a waveform, the structure of a packet 140 comprises a detection (DETECT) waveform portion 142 constituting a first waveform portion, a physical layer control (PHYCONTOL) waveform portion 144 constituting a second waveform portion, and a payload waveform portion 146 constituting a third waveform portion. The DETECT waveform portion 142 constitutes a preamble. The PHYCONTROL waveform portion 144 conveys a control code, the purpose of which is to enable a variety of data payloads to be demodulated and decoded, potentially drawing from a range of data modulation and coding formats to best fit prevalent propagation conditions. The control code informs downstream physical layer processing to switch to appropriate telemetry demodulation and channel decoding modes to recover information in payload waveform portion 146. In other examples, where combined telemetry and positioning functionality need not be supported, the structure of the packet 140 can comprise the first and second waveform portions, the second waveform portion not necessarily being a PHYCONTROL waveform portion.

[0036] Some or all of the vessels, instruments and devices of personnel of Figure 1 can be configured to operate in the underwater communications system. Turning to Figure 3, for the sake of clarity and conciseness of description, the underwater communications system will be described in simplified form in the context of a first underwater communications device 150 and a second underwater communications device 152. The first communications device 150 is capable of communicating with the second communications device 152 bidirectionally. A vessel, instrument or device of Figure 1 can comprise or carry the first communications device 150 and another vessel, instrument or device of Figure 1 can carry the second communications device 152.

[0037] The first and second communications devices 150, 152 each comprise a transceiver circuit, although in some examples a given communications device can simply comprise a receiver circuit. For the sake of clarity and conciseness of description only aspects of the underwater communications system will be described herein salient to an understanding of the core structure and functionality of the embodiments of the invention. The first and second communications devices 150, 152 each comprise a first acoustic transducer 154 and a second acoustic transducer 156, respectively, which can be employed to transmit and receive acoustic communications signals. In this example, the first and second acoustic transducers 154, 156 are acoustic transducer arrays, although in other examples single acoustic transducer elements can be employed. Any suitable transmitter chain circuit can be employed by the first and second communications devices 150, 152 to transmit underwater acoustic communications signals between the first and second communications devices 150, 152.

[0038] Referring to Figure 4, a receiver 200 of the first communications device 150 and / or the second communications device 152 comprises an acoustic transducer element 202 for receiving an acoustic communications signal 204. As indicated above, the acoustic transducer element 202 is one of a number of acoustic transducer elements of the acoustic transducer array. The acoustic transducer element 202 is operably coupled to an amplifier circuit 206, the amplifier circuit 206 being operably coupled to an analogue-to-digital converter (ADC) circuit 208. The ADC circuit 208 is operably coupled to an l / Q down-converter 210 comprising a first signal mixer 212 and a second signal mixer 214. The first and second signal mixers 212, 214 respectively comprise first inputs for receiving digital electrical signals from the ADC circuit 208 and second inputs for receiving a reference signal from a first local oscillator circuit 216. The first local oscillator circuit 216 is capable of generating the reference signal at a carrier frequency used for communication within the underwater acoustic communications system, the local oscillator circuit 216 applying an in-phase component of the reference signal to the first mixer 212 and a quadrature component of the reference signal to the second mixer 214. An output of the first mixer 212 is operably coupled to a first lowpass filter 218 and an output of the second mixer 214 is operably coupled to a second lowpass filter 220. The l / Q down-converter 210 therefore comprises an in-phase signal path and a quadrature signal path, the outputs of the first and second lowpass filters 218, 220 and hence the in-phase and quadrature signal paths are operably coupled to a complex baseband data buffer 222, constituting a part of a data store of the first and / or second communications device 150, 152, respectively. The complex baseband data buffer 222 is configured to store down-converted complex baseband data relating to the underwater acoustic communications signal 204.

[0039] In this example, the complex baseband data buffer 222 is operably coupled to a first in-phase input of a first bank of correlators 224 and a first quadrature input of the first bank of correlators 224. The complex baseband data buffer 222 is also accessible by a coarse Doppler compensator, a fine Doppler compensator, and a signal re-processor (not shown in Figure 4, but indicated by connector ‘A’). A first stored replica 226 of the DETECT waveform portion 142 is stored by another part of the data store and operably coupled to a first replica input of the first bank of correlators 224. A first output, a second output and a third output of the first bank of correlators 224 is operably coupled to corresponding inputs of a first decision unit 228, an output of the first decision unit 228 being operably coupled to a first input of a Doppler calculation unit 230 and a second input of the Doppler calculation unit 230 is operably coupled to an output of a PHYCONTROL waveform portion decoder (not shown in Figure 4, but indicated by connector ‘B’) and an output of DETECT waveform portion correlator (also not shown in Figure 4, but indicated by connector ‘C’). The Doppler calculation unit 230 has a coarse Doppler estimate, Dcoarse, output and a refined Doppler estimate, Dfine, output.

[0040] Referring to Figure 5, a coarse Doppler compensator 240 comprises a pair of quadrature inputs operably coupled to the complex baseband data buffer 222 (connector ‘A’), a first in-phase input of the coarse Doppler compensator 240 being operably coupled to a first input of a third signal mixer 242 and a second quadrature input of the coarse Doppler compensator 240 being operably coupled to a first input of a fourth signal mixer 244. Second inputs of the third and fourth signal mixers 242, 244 are operably coupled to a second local oscillator 246, the second local oscillator 246 being operably coupled to the coarse Doppler estimate, Dcoarse, output of the Doppler calculation unit 230. The second local oscillator 246 is configured to apply an in-phase component of a coarse Doppler compensation signal to the second input of the third signal mixer 242, and a quadrature component of the coarse Doppler compensation signal to the second input of the fourth signal mixer 244. The coarse Doppler compensation signal is based upon the reference signal at the carrier frequency, fc, used for communication within the underwater acoustic communications system adjusted by the coarse Doppler estimate, Dcoarse, obtained from the Doppler calculation unit 230. An in-phase output of the third signal mixer 242 and a quadrature output of the fourth signal mixer 246 are operably coupled to the PHYCONTROL waveform portion decoder 248. The PHYCONTROL waveform decoder 248 comprises a second bank of correlators 250, the in-phase and quadrature outputs of the third and fourth signal mixers 242, 244 being respectively operably coupled in-phase and quadrature inputs of the second bank of correlators 250. The in-phase and quadrature outputs of the third and fourth signal mixers 242, 244 are also operably coupled to respective in-phase and quadrature inputs of the DETECT waveform portion correlator 252. A plurality of PHYCONTROL waveform portion stored replicas 254, Ri ... R64, of the PHYCONTROL waveform portion 144 is stored by yet another part of the data store and operably coupled to a replica input of the second bank of correlators 250. The plurality of PHYCONTROL waveform portion stored replicas 254 constitute candidates for the actual symbol waveform conveyed as the PHYCONTROL waveform portion. An output of the second bank of correlators 250 is operably coupled to a second decision unit 256. An output of the second decision unit 256 is operably coupled to the second input of the Doppler calculation unit 230 as described above (with reference to connector ‘B’).

[0041] A first stored replica 226 of the DETECT waveform portion 142 stored by the data store is also operably coupled to a first replica input of the DETECT waveform portion correlator 252, an output of the DETECT waveform portion correlator 252 being operably coupled to the third input of the Doppler calculation unit 230 described above (with reference to connector ‘C’).

[0042] Turning to Figure 6, a first fine Doppler compensator 260 comprises an in-phase input and quadrature input operably coupled to the complex baseband data buffer 222 (connector ‘A’), the in-phase input of the first fine Doppler compensator 260 being operably coupled to a first input of a fifth signal mixer 262 and the quadrature input of the first fine Doppler compensator 260 being operably coupled to a first input of a sixth signal mixer 264. Second inputs of the fifth and sixth signal mixers 262, 264 are operably coupled to a third local oscillator 266, the third local oscillator 266 being operably coupled to the refined Doppler estimate, Dfine, output of the Doppler calculation unit 230. The third local oscillator 266 is configured to apply an in-phase component of a fine Doppler compensation signal to the second input of the fifth signal mixer 262, and a quadrature component of the fine Doppler compensation signal to the second input of the sixth signal mixer 264. The fine Doppler compensation signal is based upon the reference signal at the carrier frequency, fc, used for communication within the underwater acoustic communications system adjusted by the refined Doppler estimate, Dfine. The first fine Doppler compensator 260 is operably coupled to a PAYLOAD waveform portion decoder 268. The PAYLOAD waveform portion decoder 268 comprises a programmable bank of matching correlators 270 configured to calculate, in this example, signal quality indicator values. However, in other examples the signal quality indicator calculation functionality need not be employed. An in-phase input and a quadrature input of the programmable bank of correlators 270 are respectively operably coupled to an in-phase output of the fifth signal m ixer 262 and a quadrature output of the sixth signal mixer 264. The bank of matching correlators 270 comprises a payload symbol output 272 and a decoded symbol quality indicator output 274. The bank of matching correlators 270 also comprises a symbol input operably coupled to a payload symbol store 276, which in this example is part of the data store.

[0043] Referring to Figure 7, a signal re-processor comprises a second fine Doppler compensator 280 having an in-phase input and a quadrature input operably coupled to the complex baseband data buffer 222 (connector A’). The in-phase input of the second fine Doppler compensator 280 is operably coupled to a first input of a seventh signal mixer 282 and the quadrature input of the second fine Doppler compensator 280 is operably coupled to a first input of an eighth signal mixer 284. Second inputs of the seventh and eighth signal mixers 282, 284 are operably coupled to a fourth local oscillator 286, the fourth local oscillator 286 being operably coupled to the refined Doppler estimate, Dfine, output of the Doppler calculation unit 230. The fourth local oscillator 286 is configured to apply an in-phase component of the fine Doppler compensation signal to the second input of the seventh signal mixer 282 and a quadrature component of the fine Doppler compensation signal to the second input of the eighth signal mixer 284. The fine Doppler compensation signal is based upon the reference signal at the carrier frequency, fc, used for communication within the underwater acoustic communications system adjusted by the refined Doppler estimate, Dfine. The second fine Doppler compensator 280 is operably coupled to a re-processing correlator 288. The re-processing correlator 288 comprises another programmable bank of matching correlators 290 configured to calculate a correlation between known decoded symbols and fine Doppler compensated complex baseband data stored in the complex baseband data buffer 222. An in-phase input of the programmable bank of matching correlators 290 is operably coupled to an in-phase output of the seventh signal mixer 282 and a quadrature input of the programmable bank of matching correlators 290 is operably coupled to a quadrature output of the eighth signal mixer 284. The programmable bank of matching correlators 290 also comprises a replica input operably coupled to a packet symbol selector 292. The packet symbol selector 292 is configured to receive the replica of the DETECT waveform portion 142, a replica of the decoded PHYCONTROL waveform portion 144, replicas of the decoded PAYLOAD symbols 146 from the payload symbol output 272 (connector ‘D’) of the bank of matching correlators 270 of the PAYLOAD waveform portion decoder 268 along with the accompanying signal quality indicator values obtained from the decoded symbol quality indicator output 274 (connector ‘E’) of the PAYLOAD waveform portion decoder 268. In this example, the decoded PAYLOAD symbols 146 and associated corresponding signal quality indicator values are buffered by the data store after generation thereof and prior to use by the re-processing correlator 288.

[0044] A correlation output of the programmable bank of matching correlators 290 is operably coupled to a time of arrival estimator unit 294 and, in this example, a phase of arrival estimator unit 296 since an acoustic transducer array is employed. An output of the phase of arrival estimator unit 296 is operably coupled (connector ‘F’) to a bearing of arrival estimator unit 300 (Figure 8). The bearing of arrival estimator unit 300 is also operably coupled to phase angle of arrival outputs of other phase angle of arrival estimator units associated with signal processing chains configured to process acoustic signals received by other acoustic transducer elements of the array of transducer elements. The bearing of arrival estimator unit 300 comprises a bearing of arrival estimate output 302.

[0045] In operation (Figures 9 to 13), the first underwater communications device 150 transmits (Step 400) an acoustic interrogation signal to the second underwater communications device 152, for example to initiate a positioning measurement as part of the operation of a navigation subsystem of the first underwater acoustic communications device 150. The acoustic interrogation signal propagates through the water column 102 to the second underwater acoustic communications device 152, whereupon the acoustic interrogation signal is received (Step 402), demodulated and decoded. The second underwater communications device 152 then recognises that the acoustic interrogation signal is a request for an immediate response to the acoustic interrogation signal, in this example requiring the response to include a recorded time of arrival of the acoustic interrogation signal. The second underwater communications device 152 therefore generates a reply message, in the form of a data packet of the form described with reference to Figure 2, which is transmitted (Step 404) as the underwater acoustic communications signal 204 to the first underwater acoustic communications device 150. The second underwater acoustic communications apparatus 152 can employ the same techniques described herein to determine a time of arrival of the acoustic interrogation signal.

[0046] The underwater acoustic communications signal 204 transmitted by the second underwater communications device 152 is received (Step 406) by the first underwater communications device 150, the received underwater acoustic communications signal 204 having a time of arrival and a bearing of arrival associated therewith, the estimation of which (Step 408) will now be described. As it is assumed in this example that the first underwater communications device 150 comprises an array of acoustic transducer elements, the following processing is applied to the underwater acoustic communications signal 204 as received by each transducer element of the acoustic transducer array. For the sake of clarity and conciseness of description though, the processing of signals generated as a result of reception of the underwater acoustic communications signal 204 by a single acoustic transducer element will only be described until description of estimation of the bearing of arrival. As such, it should be assumed that the processing described hereinbelow applies to processing of the received acoustic communications signal 204 in respect of each acoustic transducer element.

[0047] Referring to Figures 4 and 10, the acoustic communications signal 204 is sensed by the acoustic transducer element 202 and converted (Step 420) from the acoustic domain to the electrical domain. An analogue electrical signal generated by the acoustic transducer element 202 is amplified (Step 422) by the amplifier circuit 206 before being passed to the ADC circuit 208. The ADC circuit 208 converts (Step 424) the analogue electrical signal to the digital domain prior to being down-converted (Step 426) in the IQ down-converter 210. In this regard, the digitised received passband signal is applied to both the first and second signal mixers 212, 214, and the first local oscillator 216 applies an in-phase component of the oscillating reference signal at the carrier frequency, fc, to the first signal mixer 212 and a quadrature component of the oscillating reference signal to the second signal mixer 214. The application of the in-phase and quadrature components of the reference signal to the digitised received passband signal results in the received digital passband signal being mixed down (Step 426) to baseband, the baseband digital signal having real and imaginary components, which are respectively applied to the first lowpass filter 218 and the second lowpass filter 220, which low pass filter (Step 428) the real and imaginary components of the digital baseband signal, prior to being stored (Step 430) in the complex baseband data buffer 222.

[0048] Under the control of a processing supervisor unit (not shown), the first bank of correlators 224 retrieve part of the complex baseband data stored in the complex baseband data buffer 222 corresponding to the first waveform portion of the received acoustic communications signal 204. Thereafter, the first stored replica 226 is retrieved from the data store and applied (Step 432) in multiple forms to the first bank of correlators 224, namely frequency shifted replicas of the DETECT waveform portion are applied to the first bank of correlators 224 and the first bank of correlators 224 calculates multiple correlations of the complex baseband data stored in the complex baseband data buffer 222 and corresponding to the first waveform portion of the received underwater acoustic signal 204 with the frequency-shifted replicas of the DETECT waveform portion. In one example, the output of the first bank of correlators corresponding to a correlation output having a highest peak indicative of a greatest degree of correlation can be selected, the correlation output having a frequency shift associated therewith indicative of a Doppler shift experienced by the first waveform portion of the received acoustic communications signal 204. In this example, however, a number of correlation outputs are selected (Step 434), for example correlation output signals having highest peaks of all the correlation output signals generated and the selected correlation output signals are subjected to an interpolation process to determine a true frequency at which a highest point of correlation exists. For example, a highest three correlation output signals can be selected: a first correlation output signal, a second correlation output signal and a third correlation output signal. One of the first, second and third correlation output signals represents a highest degree of correlation, a second highest degree of correlation and a third highest degree of correlation as determined by the first bank of correlators 224. In this regard, the first, second and third correlation output signal each comprise a respective first, second and third maximum value or peak. However, over the frequencies spanned by the peaks of the first, second and third correlation output signals, an optimum (higher) peak can lie between a pair of neighbouring peaks of the first, second and third correlation output signals and so by interpolating between the peaks of the first, second and third correlation output signals, it is possible to identify the true location at which maximum correlation occurs and thus the frequency at which the optimum peak resides. The Doppler shift experienced by the received acoustic communications signal can therefore be estimated coarsely, but with greater precision than by simply selecting a peak of a single correlation output signal. Consequently, the first decision unit 228 performs the interpolation (Step 436) mentioned above and from the interpolation performed identifies (Step 438) the location of maximum correlation and thus the true frequency at which maximum correlation occurs. The frequency at which the maximum correlation occurs is then passed to the Doppler calculation unit 230, which uses the frequency provided to calculate the coarse Doppler shift experienced by the received acoustic communications signal 204, for example using knowledge of the carrier frequency, fc, a difference between the carrier frequency, fc, and the location of maximum correlation calculated, and the speed of sound in water.

[0049] Once the coarse Doppler shift has been estimated, the PHYCONTROL symbol carried by the second waveform portion of the received acoustic communications signal is decoded. Referring to Figures 5 and 11, the first and second waveform portions of the received acoustic communications signal 204 are compensated for Doppler shift using the coarse Doppler shift estimated. The coarse Doppler shift estimated is applied to the parts of the complex baseband data stored in the complex baseband data buffer 222 corresponding to the first and second waveform portions of the received acoustic communications signal 204. In this regard, coarse Doppler compensator 240 obtains the complex baseband data from the complex baseband data buffer 222 corresponding to the first and second waveform portions and applies (Step 440) the reference signal adjusted by the coarse Doppler shift estimated to the complex baseband data retrieved. The part of the complex baseband data retrieved that corresponds to the DETECT waveform, namely the first waveform portion, is then buffered for correlation with the replica of the DETECT waveform portion by the DETECT waveform portion correlator 252. In this regard, the coarsely Doppler compensated version of the complex baseband data corresponding to the first waveform portion of the received acoustic communications signal 204 is correlated with the replica of the DETECT waveform portion, but not in any frequency-shifted form, as part of a process for estimating a residual amount of Doppler shift. As such, the DETECT waveform portion correlator 252 correlates (Step 442) the non-frequency shifted replica of the DETECT waveform portion with the coarse Doppler compensated complex baseband data corresponding to the first waveform portion of the received acoustic communications signal 204. The preamble correlation output signal generated by the DETECT waveform portion correlator 252 is then temporarily stored for access (connector ‘C’) by the Doppler calculation unit 230, which uses the frequency at which the peak of the output signal resides as will be described in further detail below. In order to complete the estimate of the residual amount of Doppler shift, Dres, the part of the complex baseband data corresponding to the second waveform portion of the received acoustic communications signal 204 is decoded as follows.

[0050] The part of the complex baseband data corresponding to the PHYCONTROL waveform portion, namely the second waveform portion, which has been compensated by the coarse Doppler compensator 240 is then processed (Step 444) by the PHYCONTROL waveform portion decoder 248. The PHYCONTROL waveform portion decoder 248 has access to replicas of the different forms of the PHYCONTROL waveform portion. In this example, 64 different forms of the PHYCONTROL waveform portion exist. The second bank of correlators 250 obtain each of the different possible PHYCONTROL waveform portions stored as the plurality of PHYCONTROL waveform portion replicas 254 and correlates each form of the PHYCONTROL waveform portion replica with the coarse Doppler compensated part of the complex baseband data corresponding to the PHYCONTROL waveform portion in order to determine (Step 446) which of the possible PHYCONTROL waveform portions is represented by the second waveform portion of the received acoustic communications signal 204. In this regard, the second bank of correlators 250 generates a plurality of correlation output signals, one for each replica of the different PHYCONTROL symbols. The second decision unit 256 selects the correlation output signal having the highest peak as indicative of the PHYCONTROL symbol that is conveyed by the second waveform portion of the received acoustic communications signal 204. As such, a maximum likelihood correlation process is performed and the selected correlation output signal is a maximum likelihood correlation output signal. The selected correlation output signal is then temporarily stored for access (connector ‘B’) by the Doppler calculation unit 230.

[0051] Once the second decision unit 256 has generated the maximum likelihood correlation output signal, the Doppler calculation unit 230 uses the frequency at which the peak of the preamble correlation output signal generated by the DETECT waveform portion correlator 252 resides and the frequency at which the peak of the maximum likelihood correlation output signal provided by the second decision unit 256 resides in order to calculate a difference between the two frequencies at which the respective peaks reside, thereby calculating (Step 448) the residual Doppler shift, Dres, with respect to the coarse Doppler shift previously estimated. Thereafter, the Doppler calculation unit 230 computes (Step 450) an estimate of the refined Doppler shift, Dfine, using the coarse Doppler shift, Dcoarse, already estimated and the residual Doppler shift, Dres, calculated, for example by summation of the two values. The estimate of the refined Doppler shift, Dfine, constitutes an improved estimate of the Doppler shift over the coarse estimate of Doppler shift, Dcoarse. Once the refined estimate of the Doppler shift, Dfine, has been calculated, the refined estimate of the Doppler shift, Dfine, is used to enable decoding of the PAYLOAD part 146 of the received acoustic communications signal 204 as follows.

[0052] Turning to Figure 6, the third waveform portion of the received acoustic communications signal 204 is compensated for Doppler shift using the refined Doppler shift estimated, Dfine, by applying the refined Doppler shift to the parts of the complex baseband data stored in the complex baseband data buffer 222 corresponding to the third waveform portion of the received acoustic communications signal 204. In this regard, the first fine Doppler compensator 260 obtains the complex baseband data from the complex baseband data buffer 222 corresponding to the third waveform portion and applies (Step 452) the reference signal adjusted by the refined Doppler shift, Dfine, estimated to the complex baseband data retrieved. The part of the complex baseband data retrieved from the complex baseband data buffer 222 that corresponds to the PAYLOAD waveform portion 146 and that has been fine Doppler compensated is then passed for use by the bank of matching correlators 270. The bank of matching correlators 270 correlates one symbol portion at a time of the fine Doppler-compensated complex baseband data corresponding to the third waveform portion with the replicas of possible symbols stored in the payload symbol store 276 in order to decode the PAYLOAD waveform portion of the Doppler-compensated complex baseband data. For each one symbol portion of the Doppler-compensated complex baseband data, a correlation output signal having a highest peak output value is selected and the symbol replica associated with the highest peak output value is identified by the bank of matching correlators 270 at the payload symbol output 272 as the symbol conveyed by the one symbol portion of the third waveform portion of the received acoustic communication signal 204. In addition to the identification of the most likely symbol being communicated, the bank of matching correlators 270 also generates a signal quality indicator value at the decoded symbol quality indicator output 274 indicative of the likelihood of the determination of the symbol provided at the payload symbol output 272 being correct. In this example, the signal quality indicator value is derived from a Maximum Likelihood Sequence Estimation computation, i.e. possible symbol correlations, for example by dividing a correlation peak output value of a selected symbol with a mean of the correlation peak output values of symbols that have not been selected. This process is repeated for all symbol portions of the fine Doppler-compensated complex baseband data, thereby identifying all the symbol waveforms of the third waveform portion. The symbol waveform portions of the complex baseband data identified are then demodulated and decoded (Step 454) in accordance with the control code identified above. The data conveyed by the payload portion 146 of the received underwater acoustic communications signal 204 is, in this example, encoded using an error correction coding technique, for example Forward Error Correction (FEC) coding. The symbol waveform portions are therefore decoded in accordance with the error correction technique.

[0053] Turning to Figures 7 and 12, once the symbols that have been communicated by the received underwater acoustic communications signal 204 are known at the first underwater communications device 150, the complex baseband data corresponding to the received acoustic communications signal 204 is reprocessed leveraging the a priori knowledge of the DETECT symbol, the PHYCONTROL symbol and one or more of the PAYLOAD symbols and the additional energy available to calculate at least a time of arrival of the received acoustic communication signal 204. In this regard, the complex baseband data in respect of the entire received acoustic communications signal 204 is retrieved from the complex baseband data buffer 222 by the second fine Doppler compensator 280 and Doppler compensation is applied (Step 456) to the retrieved complex baseband data using the refined Doppler shift estimate, Dtine. In this regard, an in-phase component of the reference signal at the carrier frequency, fc, adjusted by the refined Doppler shift estimate, Dtine, is applied to the in-phase component of the complex baseband data by the seventh signal mixer 282, and a quadrature component of the reference signal at the carrier frequency, fc, adjusted by the refined Doppler shift estimated, Dtine, is applied to the quadrature component of the complex baseband data by the eighth signal mixer 284. Once the complex baseband data has been compensated using the estimated refined Doppler shift, the refined Doppler-compensated complex baseband data provided to the reprocessing correlators 288 is buffered by the data store for processing in the following manner.

[0054] As the packet has been completely decoded, the content of the packet is clearly known. This a priori data can be applied to the programmable bank of matching correlators 290 in order to perform a correlation with the complex baseband data that has been fine Doppler compensated. In order to ensure the best possible result, the packet symbol selector 292 analyses the signal quality indicator values for each symbol of the third waveform portion that has been decoded and determines, for each symbol whether the signal quality indicator value conforms to a predetermined criterion, for example is of a sufficiently high value. Indeed, a threshold criterion can be applied or any other suitable criterion, for example one based upon a statistical calculation. Where a symbol of the third waveform portion has not been decoded with sufficient confidence, the symbol is omitted from reprocessing and the programmable bank of correlators 290 is dynamically configured not to correlate the relevant symbol of the third waveform portion corresponding to the symbol decoded with insufficient confidence. The packet symbol selector 290 generates a replica comprising the replica of the DETECT symbol waveform, the replica of the identified PHYCONTROL symbol waveform and the waveforms of the symbols of the PAYLOAD decoded with sufficiently high confidence.

[0055] During reprocessing, the complex baseband data that has been compensated using the refined Doppler estimate and buffered is read out in parallel from the data store and applied (Step 458) to the programmable bank of matching correlators 290 and the packet symbol selector 292 selects (Step 460) the symbol waveforms with which the fine Doppler-compensated complex baseband data is to be correlated in the manner described above. The replica symbol waveforms selected then undergo a correlation process (Step 462) with the fine Doppler-compensated complex baseband data in order to yield a reprocessing correlation output signal, which can be used to calculate an underwater navigation or positioning parameter.

[0056] The reprocessing correlation output signal is then analysed by the time of arrival estimator unit 294 and the time at which the peak of the reprocessing correlation output signal resides is used to estimate (Step 464) the time of arrival of the received acoustic communications signal, for example by searching for the peak and determining a time at which the peak occurred, which is an improvement over a time of arrival calculated using the coarse Doppler estimate. After subsequent downstream demodulation and decoding of the received underwater acoustic communications signal, the estimated time of arrival, knowledge of the transmit time of the interrogation signal mentioned earlier above, and knowledge of the turnaround time of the second underwater communications device, the first underwater acoustic communications device can calculate a round trip time from transmission of the acoustic interrogation signal to receipt of the acoustic communications signal 204. From the round-trip time, a one-way time-of-flight time, which is an example of a navigation or positioning parameter, can be calculated by a simple halving of the round-trip time calculated. The range to the second acoustic communications device 1152 from the first acoustic communications device 150 can then be calculated using knowledge of the speed of sound in the water column 102.

[0057] Additionally, the phase of arrival estimator unit 296 analyses the reprocessing correlation output signal in order to estimate (Step 466) a phase angle of arrival, for example by employing an arctan calculation using the real and imaginary components corresponding to the position of the peak in the reprocessing correlation output signal.

[0058] Referring to Figure 8 and 13, where multiple transducer elements are employed, for example in the array of transducer elements mentioned above, the phase angle of arrival estimator unit 296 communicates the estimated phase angle of arrival in respect of each transducer element of the array to the bearing of arrival estimator unit 300. The bearing of arrival estimator unit 300 determines (Step 500) whether multiple phase angle of arrival estimates are available and awaits receipt of further phase angle of arrival estimates if a sufficient number of phase angle of arrival estimates is not available. Upon receipt of a sufficient number of phase angle of arrival estimates, the bearing of arrival estimator unit 300 uses differences between multiple phase angles of arrival and knowledge of spacing between transducer elements of the array of transducer elements in order to calculate (Step 502) a bearing of arrival associated with the received acoustic communications signal 204 from the second underwater acoustic communications device 152, which is provided at the bearing of arrival estimate output 302.

[0059] The skilled person should appreciate that the above-described implementations are merely examples of the various implementations that are conceivable within the scope of the appended claims. Indeed, although the above examples have been described in the context of selecting payload symbols based upon confidence of decoding accuracy, or signal quality, the skilled person should appreciate that a predetermined number of symbols of the pay load of the packet can simply be selected for processing. Additionally or alternatively, specific payload symbol locations can be selected according to a selection criterion that does not consider quality of decoding, for example odd positional symbol locations or a predetermined number of soonest arriving symbols of the payload portion 146, for use in reprocessing the complex baseband data.

[0060] In environments or arrangements where no Doppler shift is expected to be experienced, the skilled person will appreciate that the receipt, generation of baseband data, decoding and re-processing of the baseband data need not involve the compensation for Doppler shift and the baseband data can be re-processed with the aim of improving the signal-to-noise ratio of the acoustic communications signal received.

[0061] Alternative embodiments of the invention can be implemented as a computer program product for use with a computer system, the computer program product being, for example, a series of computer instructions stored on a tangible 5 data recording medium, such as embodied in a computer data signal, the signal being transmitted over a tangible medium or a wireless medium, for example, acoustic, microwave or infrared. The series of computer instructions can constitute all or part of the functionality described above, and can also be stored in any memory device, volatile or non-volatile, such as semiconductor, magnetic, optical or other 10 memory device.

Claims

1. A method of making an improved estimate of Doppler shift by an underwater communications apparatus, the method comprising:receiving an acoustic communications signal and buffering complex baseband data derived therefrom, the complex baseband data representing a plurality of symbols organised as a first waveform portion and a second waveform portion conforming to a data structure definition;applying the first waveform portion of the complex baseband data to a first bank of correlators to identify a preamble waveform portion;calculating a first estimate of a Doppler shift in respect of the first predetermined waveform portion;coarse Doppler compensating the first and second waveform portions of the complex baseband data for the Doppler shift thereof using the first estimate of the Doppler shift;correlating the coarse Doppler-compensated first waveform portion of the complex baseband data with a replica of the first waveform portion, thereby generating a preamble correlation output signal;decoding the coarse Doppler-compensated second waveform portion of the complex baseband data, the decoding of the coarse Doppler-compensated second waveform portion comprising generating a maximum likelihood correlation output signal associated with the decoded coarse Doppler-compensated second waveform portion;calculating a residual Doppler shift using the preamble correlation output signal and the maximum likelihood correlation output signal; andcalculating the improved estimate of the Doppler shift using the calculated residual Doppler shift and the first estimate of the Doppler shift.

2. A method as claimed in Claim 1, wherein the plurality of symbols is further organised as the first waveform portion, the second waveform portion and a third waveform portion conforming to the data structure definition.

3. A method as claimed in Claim 1 or Claim 2, further comprising: decoding the coarse Doppler-compensated second waveform portion of the complex baseband data to obtain a control code.

4. A method as claimed in Claim 3, further comprising:applying the coarse Doppler-compensated second waveform portion of the complex baseband data to a second bank of correlators in order to decode the coarse Doppler-compensated second waveform portion of the complex baseband data.

5. A method as claimed in Claim 4, further comprising:calculating the maximum likelihood correlation output signal from application of the coarse Doppler-compensated second waveform portion of the complex baseband data to the second bank of correlators.

6. A method as claimed in any one of the preceding claims, further comprising: the first bank of correlators generating a plurality of correlation output signals, each of the plurality of correlation output signals resulting from correlation of a different frequency-shifted version of a replica of the first waveform portion with the first waveform portion of the complex baseband data.

7. A method as claimed in Claim 6, further comprising:selecting a first correlation output signal from the plurality of correlation output signals and in respect of a first frequency-shifted version of the replica of the first waveform portion, the first correlation output signal being indicative of a highest degree of correlation with respect to other correlation output signals of the plurality of correlation output signals;selecting a second correlation output signal from the plurality of correlation output signals and in respect of a second frequency-shifted version of the replica of the first waveform portion, the second correlation output signal being indicative of a second highest degree of correlation with respect to the other correlation output signals of the plurality of correlation output signals;selecting a third correlation output signal from the plurality of correlation output signals and in respect of a third frequency-shifted version of the replica of the first waveform portion, the third correlation output signal being indicative of a third highest degree of correlation with respect to the other correlation output signals of the plurality of correlation output signals; andcalculating the first estimate of the Doppler shift using the first, second and third correlation output signals selected.

8. A method as claimed in Claim 7, whereinthe first, second and third correlation output signals each comprise a respective first, second and third maximum value; and the calculating of the first estimate of the Doppler shift further comprises:interpolating the first, second and third maximum values to estimate an actual frequency at which maximum correlation occurs.

9. A method as claimed in Claim 8, further comprising:calculating the first estimate of the Doppler shift using the estimated actual frequency at which the maximum correlation occurs.

10. A method as claimed in Claim 3, further comprising:fine Doppler compensating the third waveform portion of the complex baseband data using the improved estimate of the Doppler shift; anddownconverting and decoding the third waveform portion of the fine Doppler-compensated complex baseband data in accordance with the control code.

11. A method as claimed in Claim 10, whereinthe data structure definition complies with an error correction technique; and the method further comprises:decoding the third waveform portion of the fine Doppler-compensated complex baseband data corresponding to the third waveform portion using the error correction technique.

12. A method of calculating an underwater navigation parameter using a received acoustic communications signal, the method comprising:calculating the improved estimate of the Doppler shift in respect of the received acoustic communication signal as claimed in any one of the preceding claims;fine Doppler compensating at least the first and second waveform portions of the complex baseband data using the improved estimate of the Doppler shift to provide fine Doppler-compensated complex baseband data; andapplying the fine Doppler-compensated complex baseband data to a bank of matching correlators, the bank of matching correlators correlating at least the first waveform portion of the fine Doppler-compensated complex baseband data with the first replica thereof and the second waveform portion of the fine Doppler-compensated complex baseband data with a second replica thereof in order generate a reprocessed correlation output signal; andcalculating the underwater navigation parameter in respect of the received acoustic communications signal using the reprocessed correlation output signal.

13. A method as claimed in Claim 12, when dependent upon Claims 2 and 10, whereinthe fine Doppler-compensated complex baseband data includes at least part of the third waveform portion of the fine Doppler-compensated complex baseband data compensated using the improved estimate of the Doppler shift; and the method further comprises:the bank of matching correlators additionally correlating the at least part of the third waveform portion of the fine Doppler-compensated complex baseband data with a third replica thereof, thereby contributing to the generation of the reprocessed correlation output signal.

14. A method as claimed in Claim 13, wherein the downcoverted and decoded at least part of the third waveform portion of the fine Doppler-compensated complex baseband data serves as the third replica of the fine Doppler-compensated baseband data.

15. A method as claimed in Claim 12 or Claim 13 or Claim 14, wherein the underwater navigation parameter is a time of arrival.

16. A method as claimed in Claim 12 or Claim 13 or Claim 14, wherein the underwater navigation parameter is a phase angle of arrival.

17. A method as claimed in Claim 13 or Claim 14, further comprising:calculating a plurality of signal quality indicator values in respect of a plurality of symbols decoded from the third waveform portion of the fine Doppler-compensated complex baseband data, respectively; andselecting a number of symbols from the plurality of symbols having respective signal quality indicator values conforming to a predetermined criterion to serve as the at least part of the third waveform portion of the fine Doppler-compensated complex baseband data.

18. A method of calculating a bearing of arrival of an acoustic communications signal, the method comprising:receiving an acoustic communications signal at a plurality of acoustic transducers; andcalculating a phase angle of arrival in respect of each of the plurality of acoustic transducers using the method of calculating the underwater navigation parameter as claimed in Claim 16; andcalculating a bearing of arrival using the plurality of phase angles of arrival.

19. A method as claimed in any one of Claims 12 to 18, further comprising:transmitting an acoustic communications interrogation signal to another underwater communications apparatus;the another underwater communications apparatus transmitting the acoustic communications signal in response to receipt of the acoustic communications interrogation signal; andrecording a time of transmission of the acoustic communications interrogation signal.

20. A method as claimed in Claim 19, when dependent upon Claim 15, further comprising:calculating a round trip time in respect of transmission of the acoustic communications interrogation signal using a time of transmission of the acoustic communications interrogation signal, a time of arrival of the acoustic communications interrogation signal, a turnaround time of the another underwater communications apparatus, and the calculated time of arrival of the received acoustic communications signal; andestimating a time-of-flight in respect of the received acoustic communications signal.

21. An underwater communications apparatus for making an improved estimate of Doppler shift, the apparatus comprising:an underwater acoustic transducer configured to receive an acoustic communications signal;a data store to buffer complex baseband data derived from the received acoustic communications signal, the complex baseband data representing a plurality of symbols organised as a first waveform portion and a second waveform portion conforming to a data structure definition;a signal processing resource comprising a first bank of correlators operably coupled to the underwater acoustic receiver;a Doppler estimator operably coupled to the first bank of correlators and configured to calculate a first estimate of a Doppler shift in respect of the first predetermined waveform portion;a Doppler compensator operably coupled to the data store and the Doppler estimator and configured to coarse Doppler compensate the first and second waveform portions of the complex baseband data for the Doppler shift thereof using the first estimate of the Doppler shift;a preamble correlator configured to correlate the coarse Doppler-compensated first waveform portion of the complex baseband data with a replica of the first waveform portion, thereby generating, when in use, a preamble correlation output signal;5 a decoder configured to generate a maximum likelihood correlation outputsignal associated with decoding the coarse Doppler-compensated second waveform portion of the complex baseband data;the Doppler estimator is configured to calculate a residual Doppler shift using the preamble correlation output signal and the maximum likelihood correlation 10 output signal; andthe Doppler estimator is configured to calculate the improved estimate of the Doppler shift using the calculated residual Doppler shift and the first estimate of the Doppler shift.15

Images