Joint low‐complexity equalization and carrier frequency offset compensation for underwater acoustic OFDM communication systems with banded‐matrix approximation at different channel conditions

Abstract

Summary Underwater acoustic wireless communication is considered as one of the most challenging communication technologies due to low propagation speed, water salinity, water depth, PH degree, and water temperature. Equalization is one of the means to improve the system performance. Matched filter (MF), zero forcing (ZF), and minimum mean square error (MMSE) equalizers can be used for channel equalization. Unfortunately, the performance of the matched filter is destroyed in the Multiple‐Input Multiple‐Output (MIMO) configurations. In addition, the ZF equalizer suffers from noise enhancement and high complexity because of the direct matrix inversion. On the other hand, the MMSE equalizer requires estimating the operating signal‐to‐noise ratio (SNR) to work properly. In this paper, we present a joint low‐complexity regularized ZF equalizer and carrier frequency offset (CFO) compensation scheme to deal with these problems. The proposed equalization algorithm mitigates the noise enhancement problem using a constant regularization parameter, and it can be implemented with low complexity using the banded‐matrix approximation. Also, we test the performance of the proposed system at different channel conditions. Simulation results show that the proposed equalizer has the ability to enhance the system performance with lower complexity than those of the other equalization algorithms at different channel conditions.

Joint low‐complexity nonlinear equalization and carrier frequency offset compensation for multiple‐input multiple‐output orthogonal frequency division multiplexing communication systems

“…At the receiver side, the CP vector is removed and the N ‐point FFT matrix F N ϵ ℂ N × N converts the time‐domain vector to frequency‐domain as: where Y ϵ ℂ 2 N × 1 is the received frequency‐domain vector, X ϵ ℂ 2 N × 1 represents the transmitted frequency‐domain samples after the mapping process, and Z denotes the noise vector. H , and η ϵ ℂ 2 N × 2 N are the frequency‐domain MIMO channel matrix, and the interference matrix, respectively, given as in References : where H ji ϵ ℂ N × N represents the frequency‐domain channel matrix between the j th transmit antenna and the i th receive antenna. The MIMO channel induces inter‐antenna interference due to its coupling nature.…”

“…Different approaches have been used to improve the performance of wireless radio communication systems such as multiple‐input‐multiple output (MIMO) configuration, Frequency‐domain equalization (FDE), and time‐domain equalization . In References , the authors presented linear equalizers that jointly perform FDE and CFO compensation with lower complexity using banded‐matrix approximation . In fact, these equalizers do not implement the successive interference cancellation (SIC) over spatial streams.…”

Performance enhancement of UWA‐OFDM communication systems based on FWHT

“…2 In general, the ocean water can be divided into four horizontal layers: the surface layer (mixed layer), seasonal thermocline layer, permanent thermocline layer, and deep isothermal layer. The UWA channel has a lot of challenges such as attenuation, water salinity, temperature, frequency-dependent attenuation, and time-varying nature.…”

“…Thus, those challenges cause a limitation to communications in sea water. 2 In general, the ocean water can be divided into four horizontal layers: the surface layer (mixed layer), seasonal thermocline layer, permanent thermocline layer, and deep isothermal layer. The sound propagation speed differs in each layer because of the variation of the water properties (salinity, pressure, and temperature).…”

“…33 Equalization is one of the most used algorithm to improve the UWA-OFDM system performance. 2,[36][37][38][39] The zero forcing (ZF) equalizer can be used for channel equalization; this type of equalizer suffers from noise enhancement. The minimum mean square error (MMSE) equalizer can mitigate the noise enhancement, but this type of equalizer suffers from high complexity and requires the estimation of the signal-to-noise ratio (SNR) to work probably.…”