Browse Prior Art Database

M/Ary Message Modulation Decoding Using an Associative Memory

IP.com Disclosure Number: IPCOM000073789D
Original Publication Date: 1971-Feb-01
Included in the Prior Art Database: 2005-Feb-23
Document File: 2 page(s) / 41K

Publishing Venue

IBM

Related People

Feder, MP: AUTHOR [+2]

Abstract

The use of an associative memory to perform the M-ary decoding process will be described. The parameters considered are: M = number of code points = size of the M-ary alphabet Upsilon = number of samples for the code point log(2)M

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 73% of the total text.

Page 1 of 2

M/Ary Message Modulation Decoding Using an Associative Memory

The use of an associative memory to perform the M-ary decoding process will be described. The parameters considered are: M = number of code points = size of the M-ary alphabet

Upsilon = number of samples for the code point

log(2)M</-Upsilon</-M

q = quantization level of each sample (number of bits/

sample)

T(s) = time between samples

H = data transmission rate.

All processing time is explicitly independent of M. The time available for these processes, T(s), is given by log(2)M over MH.

The associative memory configuration and the decode algorithm are shown in Figures A and B, respectively. The output of the process is max (

(Image Omitted)

) (sample j) and the associative memory is divided into three fields, S, B and F, to carry out this process. The S field is a series of 1's and 0's defining an error-free code point. The B field is used to store the sampled value of the input before accumulating it into the F field. The total number of associative bits required is M(log(2)Upsilon + Upsilon + 2q.).

In the algorithm shown in Figure B, the number of associative memory cycles
(i.e., read, write, search) per operation are indicated. As shown, there are two paths in the program. The left loop accomplishes the adding of samples to the code point's F field if s/j/(i) = 1. The right loop determines the maximum after Upsilon samples, outputs the code point, clears the B and F fields and loads the first sample...