A low complexity algorithm of blind estimation of convolutional interleaver parameters

Abstract

Interleaving plays an important role in many wireless communication systems, which is used to cope with burst errors and improve the reliability of data transmission. In this paper, we propose a new method to blindly estimate the parameters of convolutional interleaver in a non-cooperative context. The proposed method exploits both the property of interleaved stream and the deinterleaving technique. The product of the depth and width of the interleaver is firstly determined by analyzing interleaved data stream. Then the interleaved data stream is let to pass through a series of reconstructed deinterleavers. Thus the delay, the depth and width of the convolutional interleaver can be estimated by analyzing the deinterleaved data streams. Compared with other algorithms, the proposed method significantly reduces the searching scope and improves computational efficiency. Simulation results show that the algorithm has good performance even if errors exist.

An Improved Blind Recognition Method of the Convolutional Interleaver Parameters in a Noisy Channel

“…where µ and l are the numbers of rows and columns of R l respectively (µ l). The following conclusion has been given in [8]. There are several complete FEC coding codewords in each row of R l and the position of the codeword is fixed in each row when l ≥ BBM and l is the integer multiple of BM (shown in Figure 3).…”

“…There are several complete FEC coding codewords in each row of R l and the position of the codeword is fixed in each row when l ≥ BBM and l is the integer multiple of BM (shown in Figure 3). So, Gauss-Jordan elimination through pivoting (GJETP) [19] algorithm can be applied to find dependent columns [5], [8] in R l . When l < BBM or l is not the integer multiple of BM, dependent columns will not be found in R l .…”

“…3. Calculating the normalized rank of the lower triangular matrix and observing whether there is rankdeficiency [5], [8], [12].…”

“…Method in [6] includes a 4-dimension searching, but the algorithm is time-consuming and even difficult to realize. The algorithm is simplified in [8]. Algorithms based on rank criteria [10] are proposed in [7]- [9] to blindly estimate the convolutional interleaver parameters.…”

Blind identification of an unknown interleaved convolutional code

“…This problem has been already addressed in a series of papers [46][47][48][49][50]. All these papers assume that the interleaver is structured (a convolutional interleaver [46][47][48][49] or an helical scan interleaver [50]). It should be said here that the methods used in these papers make heavily use of the particular structure of the interleaved that is considered and some of these methods are highly sensitive to noise [47,49] or do not apply when there is noise [46,48].…”