Receiver Design for LDACS1

The design of the receiver is a crucial part of LDACS1. Basically, the receiver design is up to the implementation. Therefore,it is not define in the specification. However, if LDACS1 is deployed using an inlay approach, severe interference, mainly from DME, may occur. If no countermeasures are taken, such interference considerably degrades LDACS1 transmission performance. In view of this situation, an adapted receiver design is essential.

Figure Dme Freq Domain 

We developed an adapted receiver design which is able to cope with severe impulsive interference. This concept is presented in the following.

LDACS1 uses OFDM as modulation technique. A typical OFDM receiver structure is shown in the following figure.

Figure Ofdm Receiver

Based on the received signal, the time and frequency offset of the signal is estimated and compensated. Next, the signal is transformed into the frequency domain by means of an FFT. The transmission channel is estimated based on inserted pilot symbols. The estimated transmission channel is used to equalize the signal. Finally, the signal is demodulated and decoded to obtain estimates of the transmitted information bits. Such a structure is well known, however, prone to strong interference. This issue can be relieved by introducing appropriate interference mitigation methods as presented in the following.

To account for various interference conditions, we propose interference mitigation including time domain and frequency domain components. In addition, an iterative reception is advised. Such a structure is shown in the following figure.

Figure Ofdm Receiver Int

The conventional LDACS1 receiver is extended by the blanking nonlinearity, the frequency-selective blanking nonlinearity, the RNN equalization, and the iterative loop. These blocks are explained in the following.

  • The blanking nonlinearity blanks, i.e., sets to zero all samples of the received signal with a magnitude exceeding a predefined threshold. In such a way, impulsive interference is suppressed reliably. However, the signal blanking affects the LDACS1 signal which is a significant drawback of this scheme. What makes the blanking nonlinearity appealing is its efficiency: compared to other algorithms, the blanking nonlinearity combines a low computational complexity with a reliable mitigation of the impulsive interference, leading to a moderate to high performance gain. In addition, the blanking nonlinearity is a non-parametric approach which means that it requires no particular knowledge regarding the statistics of the interference signals.
  • A critical issue of the blanking nonlinearity is that the entire LDACS1 signal is discarded during the blanking interval despite only a fraction of the LDACS1 spectrum might be affected by interference. To relieve this issue, we introduce the frequency-selective blanking nonlinearity. It profits from combining the received signal with the signal after the blanking nonlinearity. The approach is realized by first detecting the interference at each sub-carrier using a Neyman-Pearson-like testing procedure. Provided that interference has been detected, both the received and the blanked signal are subsequently optimally combined so as to maximize the signal quality for each sub-carrier. In this way, the blanking of the LDACS1 signal is restricted to sub-carriers that are affected by impulsive interference.
  • The blanking nonlinearity induces interference between the individual sub-carriers. This inter-carrier interference limits the performance of the blanking nonlinearity. An appropriate equalization, aiming to estimate and subtract the inter-carrier interference can mitigate the detrimental influence of inter-carrier interference significantly.  We propose to adopt a recursive neural network (RNN) equalization to estimate and subtract the inter-carrier interference. The RNN equalization principle is based on estimating the inter-carrier interference by means of a non-linear decision function applied to the received data. Besides the received data, also a priori information can be incorporated into the estimation process.
  • A further performance gain can be achieved in an iterative loop. The a priori information obtained after decoding can be fed back to various receiver blocks for improving their performance. In addition, estimates of the transmission channel can further improve the performance of various receiver blocks.

By applying the presented interference mitigation algorithms and the adapted receiver structure, the LDACS1 transmission becomes robust against interference occurring in the aeronautical domain in the L-band, guaranteeing a reliable LDACS1 transmission.