A low-power, hard-decision analogue convolutional decoder using the modified feedback decoding algorithm

This paper describes the design of a constraint length 3, rate-1/2, hard-decision convolutional decoder using the modified feedback decoding algorithm (MFDA). The decoder employs mostly analogue current-mode circuits and is optimized for low-power dissipation at the expense of operating speed. This is important in many portable applications where trading size and power dissipation against operating speed is desired. Post-layout simulations using a 0.8 μm CMOS technology indicate that the decoder can operate at data rates of about 2.5 Mbit/s and dissipates only 3.9 mW from a single 3 V power supply. The layout core area is about 1 mm².