Transceivers for integrated sensing and communication

Abstract

The motivation of the work described in this Thesis is to investigate different communication waveforms, and characterise their performance when used for both communication and sensing. The need to combine communication and sensing functions in an ever more crowded spectrum has been shown by a review of the relevant published literature, in which it is concluded that this combination will be a critical consideration for the derivation and implementation of future waveforms. The trade-offs between communication and sensing performance have been identified for a range of candidate waveforms, considering applications primarily differentiated by their capability of sensing target velocity.

The work first considers the popular Orthogonal Frequency-Division Multiplexing (OFDM) as a baseline, and the relative performance of IM-OFDM. It is shown that Frequency Domain (FD) Index Modulation (IM) reduces the Time Domain (TD) Peak to Average Power Ratio (PAPR) of the OFDM signal, and increases the transmit power on the activated subcarriers through power redistribution. A novel IM-OFDM Integrated Sensing and Communication (ISAC) solution was conceived, which outperforms OFDM ISAC by collecting multiple sensing observations. The delay caused by the collection of multiple observations has no impact on the sensing performance, as the error floors remain constant when the number of collected observations is varied. For four collected observations, the IM-OFDM(4,3) scheme is shown to outperform OFDM both in terms of sensing and communication, where the parameter 3 represents MA, the number of activated subcarriers in each group, and the parameter 4 represents MG, the number of subcarriers in each group.

If a modest sensing and communication performance improvement is deemed sufficient, the IM-OFDM(4,3) configuration may be recommended. If a higher sensing performance is desired at the cost of communication performance erosion, the IM-OFDM(2,1) or IM-OFDM(4,1) schemes may be employed, depending on the target velocity and the required throughput. It has also been shown that the IM-OFDM(4,1) scheme using Phase Shift Keying (PSK) modulation has a better sensing performance than the other systems. As noted in the complexity analysis, the demodulation complexity is proportional to the number of subcarriers in a group, with IM-OFDM(2,1) having the lowest demodulation complexity amongst the IM-OFDM systems. Increasing the interpolation factor leads to a similar improvement in sensing performance in all systems, albeit at the cost of increased complexity. Increasing the number of subcarriers and the number of symbol slots will also increase the sensing resolution, but would require a substantial system modification.

Then the more recent Orthogonal Time Frequency Space (OTFS) waveform was considered as it is designed to be less affected by the Doppler shift induced by large velocities compared to OFDM. Code Division Multiple Access (CDMA) was employed to provide multi-user communication capabilities, whilst also allowing for monostatic sensing. CDMA was chosen due to its simplicity and its ability to reduce the Delay-Doppler Domain (DD) variability in the transmit signal characteristics relative to OTFS, thereby aiding sensing. This is in contrast to the other multi-user methodologies, such as the newer Sparse Code Multiple Access (SCMA) multi-user method, which increases the DD variability in the transmit signal characteristics.

Three different configurations of Code Division Multiple Access OTFS (CDMA-OTFS) were introduced. The results demonstrate that Zadoff-Chu Delay Code Division Multiple Access OTFS (Dl-CDMA-OTFS) and Delay Doppler Code Division Multiple Access OTFS (DD-CDMA-OTFS) are the configurations that consistently outperform pure OTFS sensing, whilst maintaining a similar communication performance at the same throughput. The added modulation complexity of CDMA-OTFS is similar to other OTFS multi-user methodologies, but the demodulation complexity of CDMA-OTFS is lower. CDMA-OTFS sensing can also consistently outperform OTFS sensing whilst not requiring any additional complexity for target parameter estimation. A correlation-based CDMA-OTFS sensing method was also conceived. In contrast to the data cancellation method, the correlation-based method does not allow CDMA-OTFS sensing to outperform OTFS sensing, actually resulting in an inferior performance. Hence, further work on this method was curtailed in favour of more promising schemes.

Affine Frequency Division Multiplexing (AFDM) may be viewed as a generalised form of OFDM, which can be tuned to exhibit similar characteristics to OTFS, and thus allows AFDM to mitigate the effects of higher Doppler shifts, and to separate propagation paths by their associated delays and Doppler shifts. Hence, AFDM is an attractive alternative to OTFS, especially for ISAC.

Iterative soft-Minimum Mean Square Error (MMSE) equalisation in conjunction with both Recursive Systematic Convolutional (RSC) and RSC-Unity Rate Convolutional (URC) coding has been utilised for AFDM, and this scheme is shown to exhibit a lower Bit Error Rate (BER) at high Energy per bit over Noise power (Eb /N0 ) than its equivalent OTFS counterparts, at lower matrix dimension, at high coding rates, and at low numbers of iterations. This is because AFDM possesses higher degrees of freedom than OTFS, since AFDM is a one-dimensional waveform, whereas OTFS is two-dimensional. When the number of iterations is increased, the BER performance of the AFDM configurations and their equivalent OTFS configurations are shown to be similar. At the communication receiver velocity of 150 m/s, both AFDM and OTFS tend to outperform OFDM, for both coded and uncoded transmission. Given that the RSC BER performance fails to improve beyond two iterations, this solution is recommended for low-complexity transceivers. By contrast, if the extra complexity of the RSC-URC aided transceiver is affordable, an extra Eb/N0 gain of 1.8 dB may be attained at a BER of 10^−5 and a code rate of 0.5. The sensing results show that AFDM has a comparable integer index estimation Root Mean Square Error (RMSE) performance to OTFS, with a greater sensing flexibility. Scoring the estimates by decoding the reflected data does not impact the integer index estimation, likely due to the poor resolution of the simulated systems.

More information

Identifiers

ORCID for Hugo Zakaria Jose Hawkins: orcid.org/0000-0001-5324-4795

ORCID for Lie-Liang Yang: orcid.org/0000-0002-2032-9327

ORCID for Chao Xu: orcid.org/0000-0002-8423-0342

ORCID for Lajos Hanzo: orcid.org/0000-0002-2636-5214

Catalogue record

Export record