Statistical modelling of a non‐linear high‐power amplifier with memory in multi‐input–multi‐output orthogonal frequency division multiplexing systems

Complexity‐aware‐normalised mean squared error ‘CAN’ metric for dimension estimation of memory polynomial‐based power amplifiers behavioural models

“…β is a weighting factor that is used to control the relative importance of each of the two parameters of the CAN metric. According to (8), the CAN metric is made of the NMSE which reflects the model performance, and a model complexity cost function (β • 10 log 10 (M • N )) that depends on the total number of coefficients in the model and thus its complexity. The complexity cost function definition is not unique and function monotonically increasing with the size of the model might be used in order to introduce the penalty term in the CAN metric.…”

Efficient linearisation technique for crosstalk and power amplifier non‐linearity suitable for massive MIMO transmitters

“…In multi‐branch and multi‐band transmitters, predistortion is one of the most prominent techniques for linearisation of non‐linearities and digital predistortion (DPD) technique employs the existing digital signal processor of the base station's transmitter using the concept of software‐defined‐radios [6–30]. The power consumption in implementing DPD on the digital signal processor is relatively low as compared to the high power consumption of PA.…”

“…In [19], a less complex crossover digital predistorter (CO‐DPD) has been proposed for mitigation of crosstalk and PA non‐linearity in 22 space‐frequency block code (SFBC) orthogonal frequency division multiplexing (OFDM) system. In [20], a statistical model has been proposed for a non‐linear PA in MIMO‐OFDM system. However, this model does not account for the crosstalk.…”

Modified SLM scheme of FBMC signal in satellite communications

“…In satellite communications, high power amplifiers (HPA) are used to gain sufficient transmit power. However, the linear operation range of the HPA is so limited that the distortion noise on the output signal at the HPA is significantly increased for high‐PAPR signals [7–9]. Many PAPR‐reduction methods have been proposed for multi‐carrier signals and they can be classified into three categories: traditional clipping methods [10, 11], probabilistic approaches [12, 13] and block coding techniques [14, 15].…”