Tokyo Tech News
September 30, 2010
Orthogonal frequency-division multiplexing (OFDM) is a popular method of carrying large amounts of data by layering digital signals together. It is used in many communications devices such as digital television or wireless broadband internet.
In recent years, so-called 60-GHz millimeter-wave OFDM systems have been extensively studied by dedicated researchers in the wireless personal area network (WPAN) working group of the Institute of Electrical and Electronics Engineers. Thanks to this work, single-chip 60-GHz WPAN transceivers have been developed using silicon and complementary metal oxide semiconductor (Si-CMOS) technology – widely-used materials for circuit-building - in order to reduce the transceiver sizes, power consumption, and cost.
It is, however, difficult to make a 60-GHz Si-CMOS frequency synthesizer without experiencing a relatively large amount of ‘phase noise’ which degrades the transceiver performance.
Now, Hiroshi Suzuki, Satoshi Suyama and their colleagues at Tokyo Institute of Technology have invented a ‘decision-directed phase noise compensation’ (DD-PNC) method that can drastically reduce the degradation of the OFDM transceivers due to large phase noise.
The DD-PNC method consists of three processing steps. Firstly, it generates a replica of the transmitted signal by exploiting an output of the channel decoder. Then the phase noise is estimated by applying an algorithm to the replica. Finally, the estimated amount of noise can be removed from the received signal.
The researchers found that just two iterations of the DD-PNC method in the OFDM transmission could significantly alleviate performance degradation due to large phase noise.
Given that many household devices now transmit multimedia at rates higher than one gigabit per second, wireless systems that employ the DD-PNC method on 60-GHz Si-CMOS transceivers could soon be in great demand.
Block diagram of OFDM receiver employing the DD-PNC method, which helps to remove noise from degraded signals.
Graduate School of Science and Engineering Communications and Computer Engineering