Search

This paper proposes an active radio frequency (RF) cancellation solution to suppress the transmitter (TX) passband leakage signal in radio transceivers supporting simultaneous transmission and reception. The proposed technique is based on creating an opposite-phase baseband equivalent replica of the TX leakage signal in the transceiver digital front-end through adaptive nonlinear filtering of the known transmit data, to facilitate highly accurate cancellation under a nonlinear power amplifier (PA). The active RF cancellation is then accomplished by employing an auxiliary TX chain to generate the actual RF cancellation signal, and combining it with the received signal at the receiver (RX) low-noise amplifier (LNA) input. A closed-loop parameter learning approach, based on the decorrelation learning rule, is also developed to efficiently estimate the coefficients of the nonlinear cancellation filter in the presence of a nonlinear PA with memory, finite passive isolation, and a nonlinear LNA. The performance of the proposed cancellation technique is evaluated through comprehensive RF measurements adopting commercial LTE-Advanced transceiver hardware components. The results show that the proposed technique can provide an additional suppression of up to 54 dB for the TX passband leakage signal at the LNA input, even at very high transmit power levels and with wide transmission bandwidths. Such a novel cancellation solution can, therefore, substantially improve the TX–RX isolation, hence reducing the requirements on passive isolation and RF component linearity, as well as increasing the efficiency and flexibility of the RF spectrum use in the emerging 5G radio networks. [ABSTRACT FROM PUBLISHER]

In this paper, a novel decorrelation-based concurrent digital predistortion (DPD) solution is proposed for dual-band transmitters (TXs) employing a single wideband power amplifier (PA), and utilizing only a single feedback receiver path. The proposed decorrelation-based parameter learning solution is both flexible and simple, and operates in a closed-loop manner, opposed to the widely applied indirect learning architecture. The proposed decorrelation-based learning and DPD processing can also be effectively applied to more ordinary single carrier/band transmissions, as well as generalized to more than two transmit bands. Through a comprehensive analysis covering both the DPD parameter learning and the main path processing, it is shown that the complexity of the proposed concurrent DPD is substantially lower compared with the other state-of-the-art concurrent DPD methods. Extensive set of simulation and RF measurement results are also presented, using base-station PAs as well as a commercial LTE-Advanced mobile PA, to evaluate and validate the effectiveness of the proposed DPD solution in various real world scenarios, incorporating both single-band and dual-band TX cases. The simulation and RF measurement results demonstrate excellent linearization performance of the proposed concurrent DPD, even outperforming current state-of-the-art methods, despite the significantly lower complexity. [ABSTRACT FROM AUTHOR]

In this paper, we provide detailed modeling of the spurious intermodulation distortion products appearing in the own receiver (RX) operating band as a result of coexisting transmitter (TX) and RX nonlinearities with noncontiguous carrier aggregation transmissions. Furthermore, an efficient digital front-end signal processing technique is proposed, which can flexibly mitigate the resulting RX in-band self-interference caused either by individual or simultaneously coexisting TX and RX nonlinearities. The technique is based on accurately estimating the effective leakage channel that models the nonlinearities of the TX and RX chains and the duplexer filters characteristics. In the parameter estimation stage, an observation RX chain is adopted for separately estimating the TX passband leakage response, which facilitates efficient joint estimation and regeneration of the TX- and RX-induced self-interference. In the online digital cancellation, the actual transmit data are used in conjunction with the estimated channel responses to generate a replica of the overall nonlinear self-interference, which is subsequently suppressed by subtracting it from the actual observation. In general, the proposed technique can efficiently estimate and suppress the self-interference at arbitrary spurious sub-bands located at the RX band. The performance evaluations with comprehensive numerical simulations and practical RF measurements indicate highly accurate and efficient operation, with up to 28 dB of measured self-interference suppression. [ABSTRACT FROM PUBLISHER]

Carrier aggregation transmissions in frequency division duplexing devices reduce the duplexing distance between the transmitter (TX) and receiver (RX) bands. As a consequence, the spurious intermodulation distortion products created by the nonlinear RF front-end of the TX may easily extend over to the RX band, potentially causing own RX desensitization. In this paper, we propose an efficient and computationally feasible adaptive digital identification and cancellation technique to mitigate the RX desensitization. We first show that the spurious leakage signal at own RX band depends on an equivalent leakage channel that models the overall signal leakage path including the TX nonlinearities, duplexer filter responses, and RX path. The parameters of the equivalent leakage channel can be efficiently estimated with the least squares or the recursive least squares algorithm, using the actual digital transmit data as a reference, and then used to regenerate and cancel the leakage interference from the received signal. The performance of the proposed technique is evaluated with extensive computer simulations, as well as with practical real-world RF measurements, demonstrating excellent calibration properties with up to 19-dB improvement in singal-to-interference plus noise ratio of the desired received signal. [ABSTRACT FROM PUBLISHER]