Please use this identifier to cite or link to this item:
Title: Test bed Implementation of a Novel Approach of Blind Modulation Classification
Authors: Gupta, R.
Keywords: Electrical Engineering
Issue Date: 2015
Abstract: Blind modulation classification (BMC) of a received signal is an intermediary stage between signal detection and demodulation process is the main task of a smart/adaptive receiver. The BMC is used to identify modulation format from received signal with no perception of transmitting data and many unidentified parameters at the receiver side, i.e. timing and frequency synchronization errors and multipath fading. Here we contemplate the issue of BMC in the presence of at fading and carrier frequency offset. Here the method we have proposed and implemented is for linearly modulated signals. The method is feature based (FB), i.e. a combination of both elementary and cyclic cumulant which gives an optimal performance even at low signal to noise ratio (SNR) values. The elementary fourth order cumulant is used to decide whether the constellation is from real, circular or rectangular, it is also known as macro classifier. The cyclic cumulant i.e. cyclic spectrum is used to classify modulation within a subclass of macro classifier also known as micro classifier. For the micro classification, we use different position of non-zero cyclic frequency for different modulation schemes, i.e. here we used fourth order cyclic cumulant at intermediate frequency (IF) level to classify PI/4-quadrature phase shift keying (PI/4-QPSK) and minimum shift keying (MSK) and second order cyclic cumulant has been used at baseband level to classify within a subclass of circular constellation i.e. QPSK and offset QPSK (OQPSK). The proposed method is works well in at fading channel and more robust than elementary cumulant approach. Here we have considered a six class problem which includes binary phase shift keying (BPSK), QPSK, OQPSK, PI/4-QPSK, MSK and 16-quadrature amplitude modulation (16-QAM). The objective of proposed algorithm is to work even in at fading channel as the elementary cumulant works doesn't work well in fading channel. Proposed algorithm also reduces the processing time as it needs only a short burst of data, i.e. only from two to three hundred symbols are required for successful blind estimation. Even though the availability of the spectrums is limited, huge amount of spectrum is used to transmit only redundant data or training sequences and an immense portion of the spectrum is not utilized by the primary users. The most effective way to overcome this is by designing a blind algorithm. The blind process increases the spectrum efficiency of communication systems without using redundant data or training sequences. It increases the effective data rate or freeing the spectrum for other primary users to use it. Many algorithms have been proposed based on maximum likelihood (ML) and FB i.e. higher-order elementary cumulants and cyclic cumulants of the received signal. The solution offered by the ML algorithm experience computational complexity, also very sensitive with timing and frequency offset, require a bit higher SNR. 2D histogram methods having a problem of storage and estimation of probability density function (PDF) are also difficult. The higher-order elementary cumulants method work well in additive white Gaussian noise (AWGN) channel, but this method performs poorly in at fading condition. The method we have proposed and implement is more robust than any other technique and work well even in at fading channel.
Appears in Collections:03. EE

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.