Area Efficient Sorting Unit Using Scalable Digital
B. Sargunam, S. Srinitha · 2014
Sorting is the process of arranging the data into a meaningful order so that we can analyze it more effectively. Sorting is a key requirement in many applications like digital signal processing, scientific computing, network processing etc. This paper presents an area efficient technique for designing high throughput and low latency sorting units. Two popular parallel sorting algorithms are used in this paper, they are Bitonic sorting network and odd-even merge sorting network. These sorting units utilize parallel sorting method which uses Compare-and-Exchange (CAE) blocks. When number of inputs increases, the number of CAE blocks also increases and hence the area increases. To obtain an area efficient sorting network, CAE blocks used in parallel sorting units are replaced with scalable CMOS comparators. Sorting units are coded in VHDL, simulated using Modelsim SE 10.0b and implemented in FPGA using Xilinx ISE for analysis.