Paper Info
Title
The use of interleaving to reduce the peak-to-average power ratio of an OFDM signal
Abstract
An interleaver based technique for improving the Peak-to-Average Power Ratio (PAP) of an orthogonal frequency division multiplexing (OFDM) signal is presented. For this tech- nique, K-1 random interleavers are used to produce K-1 permuted sequences from the same information sequence. PAPS of the per- muted sequences and the original information sequence are then computed using of K oversampled FFTs (OFFTs). The sequence with the lowest PAP is chosen for transmission. Complementary cumulative density function (CCDF), of PAP of an interleaved OFDM signal is observed. Results show that for 256 subcarriers and QPSK data symbols, even with K =2, the 0.1% CCDF is re- duced by 13 dB and with K4, is reduced by 2dB. The 0.1% CCDF can be reduced by 3 dB and 0.01% CCDF by 4dB at a cost of 16 OFFTs and a data rate loss of less than 0.8% with K=Z6. Further statistical improvement for the PAP is obtained by com- bining this approach and the partial transmit sequence (PTS) ap- proach. This paper also proposes an adaptive approach to reduce the complexity of the interleaving technique. frame by random vectors. In PTS, the input data frame of N symbols is partitioned into M clusters. The N/M subcarriers in each cluster (except the first cluster) are phase shifted by a con- stant factor; M-I phase factors are then selected such that the resultant PAP is minimized. Unfortunately, finding the best phase factors is a highly complex and non-linear optimization problem. Therefore some attempts have been made to reduce the complexity of the optimization: (3) has presented a subop- timal iterative flipping algorithm while (4) has derived an alter- native optimization criterion. Yet even with full optimization, the maximum O.l%PAP reduction is about 4.2dB for a 256 QPSK-carrier system. It is noteworthy that the PAP of the continuous-time OFDM signal cannot be computed (4) precisely by the use of the Ny- quist sampling rate, which amounts to N samples per symbol. In this case, signal peaks are missed and PAP reduction esti- mates are unduly optimistic. Oversampling by a factor of 4 is sufficiently accurate and is achieved by simple computing 4N- point zero-padded IDFT of the data frame. Taking a 4N length FFT of N data symbols and 3N zeros does this. We assume oversampling by a factor of 4 throughout this paper. We meas- ure the complexity of any scheme by the number of PAP com- putations required.
Year
DOI
Venue
2000
10.1109/GLOCOM.2000.891696
Global Telecommunications Conference, 2000. GLOBECOM '00. IEEE
Keywords
Field
DocType
ofdm modulation,adaptive signal processing,data communication,fast fourier transforms,quadrature phase shift keying,random processes,sequences,k-1 permuted sequences,ofdm signal,qpsk data symbols,adaptive approach,complementary cumulative density function,data rate loss,information sequence,interleaver based technique,interleaving,orthogonal frequency division multiplexing,oversampled ffts,partial transmit sequence,peak-to-average power ratio,random interleavers,statistical improvement,ofdm,par
Power ratio,Algorithm,Electronic engineering,Cumulative distribution function,Data rate,Interleaving,Orthogonal frequency-division multiplexing,Mathematics,Phase-shift keying
Conference
Volume
ISSN
ISBN
1
1930-529X
0-7803-6451-1
Citations 
PageRank 
References 
20
1.86
3
Authors
2
Name
Order
Citations
PageRank
Dhammika Jayalath119720.50
C. Tellambura214311.63