COORDINATED SCIEN CE LABORATORY
A N ITERATIVE APPROACH FOR THE
CORRECTION OF ITERATIVE ERRORS
ROBERT T. CHIEN S SE JUNE HONG
UNIVERSITY OF ILLINOIS - URBANA, ILLINOIS
AN ITERATIVE APPROACH FOR THE CORRECTION OF ITERATIVE ERRORS
by
Robert T. Chien Se June Hong
Coordinated Science Laboratory University of Illinois
Urbana, Illinois
This work was supported in part by the Joint Services Electronics Program (Uo S» Army, U.S. Navy, and U.S. Air Force) under Contract DAAB-07-67-C-0199, and in part by the National Science Foundation under Grant No. GK-2339.
Reproduction in whole or in part is permitted for any purpose of the United States Government.
This document has been approved for public release and sale; its distribution is unlimited.
Other CSL Reports in Information Science include:
1» Kasami, Tadao, "An Efficient Recognition and Syntax-Analysis Algorithm for Context-Free Languages," March, 1966, R-257.
2» Barrows, J. T., Jr,, "A Topological Technique for Analysis of Active Networks," August, 1965, R-266.
3. Barrows, J, T., J r „, "A New Method for Constructing Multiple Error Correcting Linear Residue Codes," January, 1966, R-277.
4, Lum, Vincent, "A Theorem on the Minimum Distance of BCH Codes over GF(q)," March, 1966, R-281.
5. Preparata, Franco P„, "Convolutional Transformations of Binary Sequences: Boolean Functions and Their Resynchronizing Properties, March, 1966, R-283o
6. Kasami, Tadao, "Weight Distribution Formula for Some Class of Cyclic Codes," April, 1966, R-285»
7= Kasami, Tadao, "A Note on Computing Time for Recognition of Languages Generated by Linear Grammars," April, 1966, R-287.
8o Lum, Vincent, "On Bose-Chaudhuri-Hocquenghem Codes over GF(q)," July, 1966, R-306,
9. Bahl, Lalit Rai, "Matrix Switches and Error Correcting Codes from Block Designs," August, 1966, Thesis, R-314«
10« Kasami, Tadao, "Weight Distributions of Bose-Chaudhuri-Hocquenghem Codes," August, 1966, Thesis, R-317.
H o Preparata, F - P., Metze, G., and Chien, R. T., "On the Connection Assignment Problem of Diagnosable Systems," October, 1966, R-322.
12 » Chien, R„ T. and Preparata, F. P , , "Topological Structures of Information Retrieval Systems," October, 1966, R-325o
13o Heller, James Ernest, "Decoding Procedures for Convolutional Codes," November, 1966, R-327»
14. Lum, Vincent and Chien, R= T,, "On the Minimum Distance of Bose- Chaudhur i-Hocquenghem Codes," November, 1966,
R-328-/
CSL Reports (continued)
Hsu, Hsung Tsao, "Error Correcting Codes for Compound Channels," December,
1966, R-331o
16o Hong, Se June, "On Minimum Distance of Multiple Error Correcting Arithmetic Codes," January, 1967, R-336.
Gaddess, Terry G., "A Study of an Error Detecting Parallel Adder," January, 1967, M.S. Thesis, R-337.
18o Preparata, Franco P., "Binary Sequence Convolutional Mapping: The Channel Capacity of a Non-Feedback Decoding Scheme, March, 1967, R-345.
19. Preparata, F. P. and Chien, R. T., "On Clustering Techniques of Citation Graphs," May, 1967, R-349.
Lipovski, Gerald J . , "Compatibility and Row-Column Minimization of Sequential Machines," May, 1967, R-355.
2 1
.
Lipovski, Gerald J., "An Improved Method of Finding all Largest Combinable Classes," August, 1967, R-362.22. Chow, David K . s "A Geometric Approach to Coding Theory with Application to Information Retrieval," October, 1967, R-368.
23. Tracey, Robert J., "Lattice Coding for Continuous Channels," December, 1967, R-371.
24. Preparata, Franco P., "A Study of Nordstrom-Robinson Optimum Code," Aprii, 1968, R-375.
25. Preparata, Franco P., "A Class of Optimum Nonlinear Double- Error-Correcting Codes," July, 1968, R-389.
26. Kisylia, Andrew Philip, "An Associative Processor for Information Retrieval," August, 1968, R-390.
27. Beach, Edward J., "A Study of a Feedback Time-Sharing System," September, 1968, R-391.
28. Weston, P., and Taylor, S. M., "Cylinders:: A Data Structure Concept Based on Rings," September, 1968, R-393.
CSL Reports (continued)
29» Bahlj Lalit Rai, "Correction of Single and Multiple Bursts of Error," October, 1968, R-397.
3 0
. Carroll, D. E., Chien, R. T., Kelley, K. C. Preparata, F. P., Reynolds, P. , Ray, So R. and Stahl, F. A., "An Interactive Document Retrieval System," December,1968, R-398o
31o
Biss, Kenneth, "Syntactic Analysis for theR2
System," December,1968, R-399o
3 2 o Tzeng, Kenneth Kai Ming, "On Iterative Decoding of BCH Codes and Decoding Beyond the BCH Bound," January, 1969, R-404. 33
o
Kelley, K.Co,
Ray,So
R. and Stahl, F. A., "ISL-A StringManipulating Language," February, 1969, R-407.
34o
Lombardi, Daniel Joseph, "Context Modeling in a Cognitive Memory," February, 1969,R-408o
35o
Chien, R. T., Hong, S.Jo
and Preparata, F. P . , "Some Results in the Theory of Arithmetic Codes," May, 1969, R-417.3 6 o Hong, SeJune, "On Bounds and Implementation of Arithmetic Codes," October, 1969, R-437»
37» Chien, R. To and Hong, So J», "Error Correction in High Speed Arithmetic," October, 1969, R-438.
3 8 o Chien, R. T. and Hong, S. J., "On a Root-Distance Relation for Arithmetic Codes," October, 1969, R-440.
39o
Chien, R„To,
"Recent Developments in Algebraic Decoding," November, 1969, R-441»For copies of these reports please complete this form and send to Professor R, To Chien
Coordinated Science Laboratory University of Illinois
Urbana, Illinois 61801 My name and address is ________________
AN ITERATIVE APPROACH FOR THE CORRECTION OF ITERATIVE ERRORS*
Robert To Chien Se June Hong
Coordinated Science Laboratory University of Illinois
Urbana, Illinois
This work was supported principally by the National Science Foundation under Grant GK-2339 and, in part, by the Joint Services Electronics Program under Contract DAAB-07-67-C-0199»
2
ABSTRACT
The errors most likely to occur in a high-speed multiplier are called the iterative errors. An arithmetic coding technique for the correction of such error patterns is proposed. We present a class of codes and show its error correcting ability. The unique feature of this code is an iterative decoding method.
3
I. INTRODUCTION
The high-speed multiplier schemes such as the one proposed by MacSorley [ l] have been well investigated and implemented in many computers.
In such a multiplier scheme, the multiplier is divided into blocks of two (or more) bits each and each block is multiplied to the multiplicant to form partial products. The partial products are then appropriately shifted and added in a multi-input parallel adder with minimum carry provisions. The expected error pattern is quite different from either the multiple
independent errors or the burst errors. These errors are shown to be iterative in nature and of the following special form. We let m = the length of a block in bits, r = the number of blocks, and let E be a single iterative error.
k r ~1 mi
Definition 1 E = + 2 £ e.2 , where 0 < k < m and e. = 0 or 1 for all i.
i=0 1 L
A large class of arithmetic codes for the correction of such errors has been developed by Chien and Hong [2,3]. It has been shown that this class of codes has an easy implementation scheme and a nearly optimal
rate. We propose a different class of arithmetic codes here, which is based on the concept of an iterative decoding method for the iterative errors.
Arithmetic codes are designed to detect or correct errors in
digital computations. One such error may change many output digits by propa gations. Single error correcting codes are summarized in Peterson [4], and multiple independent error correcting codes have been studied by Barrows [5], Mandelbaum [6], Chang and Tsao-Wu [7] and Chien, Hong, and Preparata [8,9].
Burst error correcting arithmetic codes have been investigated by Stein [10], Chien [11], and Mandelbaum [12].
4
Arithmetic codes are of the form AN, where A is a fixed integer called the generator. N is an integer in the interval (0, B-l), and B is the number of code words. If the code length is n, B is the smallest integer such that AB>2 . In the binary case, A is obviously an odd number. The error correcting capability of ordinary AN codes depends on the minimum distance of the code, which in turn depends on the generator A. A corrupted
signal (correct signal plus error) modulo A is called the syndrome of the error which is the same as the error modulo A. Syndrome of an error, usually denoted as S, then leads to the correct decision of the error through the decoding algorithms.
II. DERIVATION OF THE CODE
It follows from the definition that to correct the error one must correctly determine the polarity of the error, the position of the error (k) and the distribution of the erroneous digits, i.e., the set of e ^ s . The class of codes dealt with in this work is for the cases when the number of
tl
blocks, r, is two to the some power, i.e., r = 2 for some t^ > 1. Note that the length of the code is m2 and 2 -1 is now divisible by 2m + 1 for all 0 < i < t^-1.
The Polarity of Error
m2
Let t„ be some integer less than t,. Consider the positive error -1. Clearly
ti tn
, 2 l-l . ,2 °-l . > _
E = E'= +2k £ e.2mi = 2k £ f .2mi mod 2™2 _1
i=0 1 i=0 1
5
t -1 t
where 0 < < 2 for all 0 < i < 2 ^-1. Thus, each f can have at the most (t^-tg)l's in its binary form. If t^-t^ < -|m, the whole residue
t0~ ^
must have less than (2 m) 1's .
Lemma 1 Given tn > t,- — m f- U
1
z.m2 0
V
1(
1
)S = E mod 2 -1 has less than (2 m) l's if and only if the polarity of error is positive.
Proof We must show that when the polarity is negative, S has greater than
t0~1 m2fc0
(2 m) l's. Let E = - E 1 and S' = E 1 mod 2 -1. We know that S' has to~1
less than (2 m) l's. Therefore,
„m2 , ,
S = 2 -1 -S'
0
V
1V
1and the number of ones in S is greater than m2 - (m2 ) = m2 . W e mention here that S = 0 only if E = 0, i.e., no error.
Q.E.D
Intermediate Error Pattern
Using the same notation as E = + E', or
E ' = 2k I e.2ml i=0 L
(
2
)we now define an intermediate error pattern as
£ . = E ' mod 2m2 ~l-1
6
for all tQ < j < t^. Clearly, £ fc = E , and from Eq. (2) we have
V 2 j'1
£ . = 2k S a.2mi (4)
J i=0 L
t l"j m 2 J
where 0 < a. < 2 and 0 < k < m. Also, note that £. = £ , mod 2 -1
“ 1 “ " J J+l
for all j < t ^ .
Consider an intermediate error pattern, £ ^ , given in an ordinary binary form. Each a^ becomes a burst of length at most t^-j with at least m ~(b-^"j) 0's in between. These bursts can be uniquely recognized if
integer such that tf-j < "2 m, i.e., if j > t^- m. Let k. be the maximum
k i tr j J
2 J a^ < 2 for all i, for the given £^ of Eq. (4). C
Now denote by £ . the following equation which is numerically the same as £
J j Clearly, k . > 0. J “
2J-1
£ . = 2^k k j^ 2 (a 2 J)2mi 3 i=0 1 (5)Lemma 2 If t^ > t^- — m, £^ can be uniquely determined from the binary pattern of £„, for all t„ < j < t,-l.
Proof tQ > t1- j m implies j > t ^ | m for all given j's. Thus, the
bursts of a ^ ’s are uniquely recognized for all j. Now mark the position of . ..
[y] th bit after the longest burst and each m th bit positions thereafter,
The term, "burst", denotes a binary pattern beginning and ending with l's A single 1 is considered as a burst of length one. A cyclic connection between m 2 J th bit and the first bit is assumed.
7
.m2'
cyclically around the entire length of 2 bits. These marks fall among the 0's separating the bursts. Let the position of the smallest marked bit be k / , we have
, /2“'-l , . / j
2k E (a.2k 'k )"2mi s £ . mod 2m2 -1
i=0 1 J
/ k - k 7 “1 J
Now, change k until (a/2 ) < 2 for all i, for the first time. By the definition of k ., k' = k - k . and k-k' = k. for Eq. (5). We mention here
k] J J
that any time (a^2 J) becomes an odd number, k^ = 0 and the position of the error, k = (k-k.).
j
Q.E.D. ti-J
Suppose a binary pattern of £ is given and (k-k ) and (a^2 J) 's are all decided according to lemma 2. We let
2^+ -l
l' = 2 (k-k j> S b' 2mi
J+1 i=0 L
(6)
t,-j-l k.
where 0 < b. < 2 and b' + b' _j = (a.2 for all i.
- i - l i+2 i
k 2j+“1 • t -j-k.
Lemma 3 Let £ = 2 E b.2m i . 0 < b. < 2 J for all 0 < i < 2 - 1
J i=0 1 1
t -j-1
Proof From Eq. (4), we know that 0 < b < 2 for all i. Also, from the definition of £., a. = b. + b.,_4 for all 0 < i < 2 J-1. Now, since
i . . J i l, i+2J . — — , *
k i V J k i t,-j
(a^2 ) < 2 , (b^ + b^+
2
j) £ 2 . Thus, b^ < 2 ^ regardless of k ..J
8
The g-Code
Define a class of integers, 3., as the following, p . is a prime
m2 j J J
factor of 2 +1, such that x = m2^ is the least positive solution for
X
2 +1 = 0 mod P
y
P^ is said to have order n if2 j+-l
2 6.2 ' ^ 0 mod 2 + 1 implies E e.2 0 mod 3
i=0 1 i=0 1 j
where e. - 1 or 0 for all i. An equivalent condition is
2J-1
miE a .2 ^ 0 mod P .
i=0 1 J
(7)
n_
where |a^ j < 2 for Finding the problem. But one can shows a short list of order at least one.
all i and not all a.'s are 0, l
order of given p seems to be a difficult number theory easily find the order by a computer programming. Table 1
P ’s and the orders. It appears that all the P .'s have J Table 1. P . and o r d e r . m j 3 1 13 3 2 241 5 1 41 5 2 61681 6 1 241 6 2 673 order m j order 2
2
28
82
6 4 97 2 6 8 193 1 7 2 29 3 7 2 113 4 7 4 1579031 5 7 8 5153 19
For a given in and r = 2 \ the 3 -code is defined under the follow ing assumptions , i) > tQ > m and tQ > 0; ii) there exist p 's (t^ < j < t^-2) of order at least t^-j+2 and |3 t ^ of order 2. When such 3 j's exist we define the generator of the (3-code as
AP = (21”2 -l)Pt0 P to+i ••• P V 1 (8)
We mention here that this generator divides 2mr-l and therefore resembles the form of the generators for ordinary multiple error correcting arith metic codes [5-10].
III» ITERATIVE DECODING
The decoding is done by iteratively determining the intermediate error patterns» We first show how is obtained from given £. and present
j+1 J
the complete decoding algorithm. An example follows for illustration.
Lemma 4 Assume the order of 3 is greater than or equal to t^-j+2,
s 8 . mod (3 . if and only if £ = £' , for all t_ < j < t--l.
J+1 J+1 J y j+1 j+1 o - J - 1
Proof We must show that £ . = £ [ , mod 3 . implies £ , = £ ; , . Now,
--- ---
J+1
J+1
Jj+1
j+1
£ = 2^~^j & j+l 2 J 2 3+-l k . k-k. 2 J+-1 E b.2 J 21"1 = 2 3 E b' 2mi mod P . i=0 1 i=0 1 3 2 j+-l Ei=0
(b.2 3-b/.)2mL = 0 mod P . or10
but
2mi „ _2m(i+2J) mQd
Thus 2 J -1
S f (bi_b±H-2 2 J " (b * ~ b ^_,_o J) 1 2mL = 0 mod 3
i=0 i+2
Since I(b £_b i+2j)| - 2 1 ^ by lemma 3 and | b^ - b[. ?j\ < 2 1 ,
ir k. t,-j t.-j-l tt-i+1
l{(bi'b i+2J)2 J - <b i - b i+ 2 j) 5 l - 2 + 2 < 2 ’ for all j. By
the definition of 3 > J
But
{bi-b l+ 2 j)2 J ' (bi'b i+23) ' 0
(bi+ b i+2j)2 J = a.2 J = (b-+b h 2 j)
Therefore b' = b^2 J for all 0 < i < 2^+ -l
Q.E.D
Theroem 9 The 3-codes, when exist, correct all single iterative errors«
Pro°f Let the initial syndrome be = A^N + E = E mod .
m2 ^ tQ 1
Step 1) If h(S mod 2 -1) < m2 , the polarity is positive, and other
wise negative« (By lemma 1.) If positive S^ = Sq, and if negative Sf = Ap - SQ „ In either case S^^ = E ; mod A^ .
p = _m2 0
~ L “ b i mod 1 ~l° However, the £ obtained now is in binary
0 c0
pattern. Iteratively follow the next step for t <i<t -2 0— — 1 Step 2)
11
/
Step 3) From the binary £., f ind £ .by lemma 2« Using St = E . mod 3., find
J “ J 1 J J
£ uniquely from £^ by lemma 4.
i'2tl“il
K ^ rn 1
Step 4) Let £ n = 2 £ a „2 „ i) If S, = 0 mod 3 -, then a. = 0
t,-1 „ A i 1 tn- l ’ l
1 i=0 1
or 2 for all i and k = k' „ ii) If S^ f 0 mod 3 fc and a, = 0 or 2 for all i, then k = k' +1. iii) If S ^ 0 mod 3 fc ^ and a^ = 0, 1, or 2 for all i, then k = k"
t,-l Step 5) Let £ = £ 1 = 2 E a .2 tl“ i i=0 L „k-k „mi Let £ 2 tl"1i ' = 2k 2 _1 e'. 2mi h 1=0
(9)
where 0 < e[ < 1 for all i and e*+e , , = a.2k k for all i.
" 1 _ 1 1
By the same arguments as lemma 4, £ # = £ = E if and only if t l tl
£" = £ mod 3. " Q.E »D.
ti
*=1V
1Example Let m = 6» Table 1 gives 3 = 241 with order 8 and 32 = 673 with order 2» Let t^ = 2, i.e», r = 8. t^ = 1 satisfies the condition for 3 -code, thus
Ap = (212-1)*241*673
the rate of which is approximately 0.4= Suppose the error is (e^,e^,...,e^)
12
(10110101), k = 3 and of positive polarity. £^ - E* mod 2 -1 becomes the following binary pattern with the marks,
T. ([f]
= 3)12
0 1 2 3 4 5 6 7 8 9 10 11 - binary positions
0 0 0 0 1 0 0 0 0 1 1 0 — > (11) is the longest burst
1 first mark jsecond mark - marking
k' = l, (a.2k 'k ) = 8, (a 2k 'k ) = 12 k ^ k (k-k2 ) = 3, (aj2 2 ) = 2, (a22 2 ) = 2 — 4 applying lemma 2 & 1 = 23 (2 °2° + 3 °26) —> Eq. (5) 0<b'<2s b'+b' = 2, bj+b' = 3 - for Eq. (6)
& 2 = 23 (l°26 0 + l “26oI+ l “26 °2+2-26 3 ) - by lemma 4 Si = Ei # 0 mod 673, thus k = 3 —> by step 4.
e > ° ; e0+ e 4= e l+ e 5=e2+ e 6=e3=e7=1 - E q . (9) , Step 5 8^ = mod 673 when (eQ,e|,. . .,e^) = (10110101) . Thus decoded.
IV. CONCLUSION
The 3 -codes are based on an interesting decoding method, namely, an iterative decoding for the iterative errors. From the syndrome of an iterative error, intermediate error patterns are iteratively decoded, each time doubling the length of the pattern. Although some searching and matching operations are necessary at each step, the unusual feature of this decoding technique may be desired for some applications.
The 3 -codes of high rate do not seem to exist for small m's. H ow ever, for large m, it is very probable that such P 's exist. The rate of the 3 -code is generally less than the rate of the codes described in [3]. Again,
13
The decoder design, a theory or a simple method to find the order of 3 and a proof of existence of 3 -codes for large m are interesting
problems for further research. Also, the iterative decoding concept may find a useful application in the polynomial codes.
V. REFERENCES
[ 1] MacSorley, O.L., "High-Speed Arithmetic in Binary Computers," Proc. of IRE, Vol. 49, No. 1, January, 1961.
[2] Hong, S„J., "Arithmetic Codes for High Speed Multipliers," Ph.D. Thesis, University of Illinois, Urbana, Illinois, July, 1969.
[3] Chien, R.T. and Hong, S.J., "Error Correction in High Speed Arithmetic," Submitted for publication in Information and Control, August, 1969.
[4]
Peterson, W . W . , "Error Correcting Codes," The M.I.T. Press, Cambridge, Massachusetts, 3rd Ed., July, 1965.[5] Barrows, J.T., Jr., "A New Method for Constructing Multiple Error Correcting Linear Residue Codes," Report R-277, Coordinated Science Laboratory, University of Illinois, Urbana, Illinois, January, 1966. [6] Mandelbaum, D., "Arithmetic Codes with Large Distance," IEEE Trans,
on Information Theory, Vol. IT-13, No. 2, April, 1967.
[7] Chang, S.H. and Tsao-Wu, N.T., "Discussion on Arithmetic Codes with Large Distance," IEEE PGIT-14, January, 1968.
[8] Chien, R.T., Hong, S.J. and Preparata, F. P., "Some Contribution to the Theory of Arithmetic Codes," Proceedings of the First Annual Hawaii International Conference on Systems Sciences, January, 1968. [9] Chien, R.T., Hong, S.J. and Preparata, F.P., "Some Results in the
Theory of Arithmetic Codes," submitted for publication, Report R-417, Coordinated Science Laboratory, University of Illinois, Urbana,
Illinois, May, 1969.
[10] Stein, J.J., "Prime Residue Error Correcting Codes," IEEE Trans, on Information Theory, Vol. IT-9, July, 1962.
[11] Chien, R.T., "Linear Residue Codes for Burst Error Correction," IEEE Trans, on Information Theory, Vol. IT-10, April, 1964.
[12] Mandelbaum, D., "Arithmetic Error Detecting Codes for Communication Links Involving Computers," IEEE Trans, on Communication Technology, Vol. Com. 13, June, 1965.
Distribution List as of
1 October1969
Dr A.A. Dougal Asst Director (Research) Ofc of Defense Res & Eng Department of Defense Washington, D.C. 20301 Office of Deputjr Director (Research and Information, Rm 3D1037) Department of Defense
The Pentagon Washington, D.C. 20301 Director.Advanced Research Projects Agency
Department of Defense Washington, D.C. 20301 Director for Materials Sciences Advanced Research Projects Agency Department of Defense Washington, D.C. 20301 Headquarters
Defense Communications Agency (340) Washington, D.C. 20305 Defense Documentation Center Attn: DDC-TCA Cameron Station
Alexandria, Virginia 22314 (50 Copies) Director
National Security Agency Attn: TDL
Fort George G. Meade, Maryland 20755 Weapons Systems Evaluation Group Attn: Colonel Blaine 0. Vogt 400 Army-Navy Drive Arlington, Virginia 22202 Central Intelligence Agency Attn: OCR/DD Publications Washington, D.C. 20505 Hq USAF (AFRDD) The Pentagon Washington, D.C. 20330 Hq USAF (AFRDDG) The Pentagon Washington, D.C. 20330 Hq USAF (AFRDSD) The Pentagon Washington, D.C. 20330 Colonel E.P. Gaines, Jr. ACDA/F0
1901 Pennsylvania Ave N.W. Washington, D.C. 20451 Lt Col R.B. Kalisch (SREE) Chief, Electronics Division Directorate of Engineering Sciences Air Force Office of Scientific Research Arlington, Virginia 22209 Dr I.R. Mirman AFSC (SCT)
Andrews Air Force Base, Maryland 20331 AFSC (SCTSE)
Andrews Air Force Base, Maryland 20331 Mr Morton M. Pavane, Chief
AFSC Scientific and Technical Liaison Office 26 Federal Plaza, Suite 1313
New York, New York 10007 Rome Air Development Center Attn: Documents Library (EMTLD) Griffiss Air Force Base, New York 13440 Mr H.E. Webb (EMMIIS)
Rome Air Development Center Griffiss Air Force Base, New York 13440 Dr L.M. Hollingsworth
AFCRL (CRN) L.G. Hanscom Field Bedford, Massachusetts 01730 AFCRL {¡RMPLR), Stop 29 AFCRL Research Library L.G. Hanscom Field Bedford, Massachusetts 01730
Hq ESD (ESTI) L.G. Hanscom Field
Bedford, Massachusetts 01730 (2 copies) Professor J. J. D'Azzo
Dept of Electrical Engineering Air Force Institute of Technology Wright-Patterson AFB, Ohio 45433 Dr H.V. Noble (CAVT) Air Force Avionics Laboratory Wright-Patterson AFB, Ohio 45433 Director
Air Force Avionics Laboratory Wright-Patterson AFB, Ohio 45433 AFAL (AVTA/R.D. Larson Wright-Patterson AFB, Ohio 45433 Director of Faculty Research Department of the Air Force U.S. Air Force Academy Colorado Springs, Colorado 80840 Academy Library (DFSLB) USAF Academy
Colorado Springs, Colorado 80840 Director
Aerospace Mechanics Division Frank J. Seiler Research Laboratory (OAR) USAF Academy
Colorado Springs Colorado 80840 Director, USAF PROJECT RAND Via: Air Force Liaison Office The RAND Corporation Attn: Library D 1700 Main Street Santa Monica, California 90045 Hq SAMS0 (SMTTA/Lt Nelson) AF Unit Post Office Los Angeles, California 90045 Det 6, Hq OAR
Air Force Unit Post Office Los Angeles, California 90045 AUL3T-9663
Maxwell AFB, Alabama 36112 AFETR Technical Library (ETV.MU-135) Patrick AFB,Florida 32925 ADTC (ADBPS-12) Eglin AFB, Florida 32542 Mr B.R. Locke
Technical Adviser, Requirements USAF Security Service Kelly Air Force Base, Texas 78241 Hq AMD (AMR)
Brooks AFB, Texas 78235 USAFSAM (SMKOR) Brooks AFB, Texas 78235 Commanding General
Attn: STEWS-RE-L, Technical Library White Sands Missile Range New Mexico 88002 (2 copies) Hq AEDC (AETS)
Attn: Library/Documents Arnold AFS, Tennessee 37389 European Office of Aerospace Research APO New York 09667
Phsical & Engineering Sciences Division U.S. Army Research Office 3045 Columbia Pike Arlington, Virginia 22204 Commanding General U.S. Army Security Agency Attn: IARD-T Arlington Hall Station Arlington, Virginia 22212
Commanding General U.S. Army Materiel Command Attn: AMCRD-TP Washington, D.C. 20315
Technical Director (SMUFA-A2000-107-1) Frankford Arsenal
Philadelphia, Pennsylvania 19137 Redstone Scientific Information Center Attn: Chief, Document Section U.S. Army Missile Command Redstone Arsenal, Alabama 35809 Commanding General U.S. Army Missile Command Attn: AMSMI-REX
Redstone Arsenal, Alabama 35809 Commanding General
U.S. Army Strategic Communications Command Attn: SCC-CG-SAE
Fort Huachuca, Arizona 85613 Commanding Officer
Army Materials and Mechanics Res. Center Attn: Dr H. Priest
Watertown Arsenal Watertown, Massachusetts 02172 Commandant
U.S. Army Air Defense School Attn: Missile Science Division, C6S Dept P.0. Box 9390
Fort Bliss, Texas 79916 Commandant
U.S. Army Command & General Staff College Attn: Acquisitions, Library Division Fort Leavenworth, Kansas 66027 Conmanding Officer
U.S. Army Electronics R&D Activity White Sanda Missile Range, New Mexico 88002 Mr Norman J. Field, AMSEL-RD-S Chief, Office of Science & Technology Research and Development Directorate U.S. Army Electronics Command Fort Monmouth, New Jersey 07703 Commanding Officer Harry Diamond Laboratories Attn: Dr Berthold Altman (AMXD0-TI) Connecticut Avenue and Van Ness St N.W. Washington, D.C. 20438
Director
Walter Reed Army Institute of Research Walter Reed Army Medical Center Washington, D.C. 20012 Commanding Officer (AMXRD-BAT) U.S. Army Ballistics Research Laboratory Aberdeen Proving Ground
Aberdeen, Maryland 21005 Technical Director U.S. Army Limited War Laboratory Aberdeen Proving Ground Aberdeen, Maryland 21005 Commanding Officer Human Engineering Laboratories Aberdeen Proving Ground Aberdeen, Maryland 21005 U.S. Army Munitions Command Attn: Science & Technology Br. Bldg 59 Picatinny Arsenal, SMUPA-VA6 Dover, New Jersey 07801 U.S. Army Mobility Equipment Research and Development Center
Attn: Technical Document Center, Bldg 315 Fort Belvoir, Virginia 22060 Director
U.S. Army Engineer Geodesy, Intelligence & Mapping Research and Development Agency Fort Belvoir, Virginia 22060 Dr Wrrwu It obi Deputy Chief Scientist U.S. Army Research Office (Durham) Box CM, Duke Station Durham, North Carolina 27706
Richard 0. Ulsh (CRDARD-IPO) U.S. Army Research Office (Durham) Box CM, Duke Station Durham, North Carolina 27706 Mr Robert 0. Parker, AMSEL-RD-S Executive Secretary, JSTAC U.S. Army Electronics Command Fort Monmouth, New Jersey 07703
Naval Electronic Systems Conmand E1£X 03, Room 2046 Munitions Building Department of the Navy Washington, D.C. 20360 (2 copies) Commander
Naval Electronics Laboratory Center Attn: Library
San Diego, California 92152 (2 copies)
Conmander, U.S. Naval Missile Center Point Mugu, California 93041 W.A. Eberspacher, Associate Head Systems Integration Division Code 5340A, Box 15 U.S. Naval Missile Center Point Mugu, California 93041
Commanding General U.S. Army Electronics Command Fort Monmouth, New Jersey 07703 Attention:AMSEL-SC RD-GF RD-MT XL-D each sym bol listed individu ally addressed XL-E XL-C XL-S (Dr R. Buser) HL-CT-DD HL-CT-R HL-CT-L (Dr W.S. McAfee) HL-CT-0 HL-CT-I HL-CT-A NL-D NL-P NL-P-2 (Mr D. Haratz) NL-R (Mr R. Kulinyi) NL-S KL-D KL-S (Dr H. Jacobs) KL-SM (Drs Schiel/Hieslmair) KL-T VL-D VL-F (Mr R.J. Niemela) WL-D
Dr A.D. Schnitzler, AMSEL-HL-NVII Night Vision Laboratory, USAEC0M Fort Belvoir, Virginia 22060 Dr G.M. Janney, AMSEL-HL-NVOR Night Vison Laboratory,USAEC0M Fort Belvoir, Virginia 22060 Atmospheric Sciences Office Atmospheric Sciences Laboratory White Sands Missile Range New Mexico 88002 Missile Electronic Warfare, Technical Area, AMSEL-WT-MT White Sands Missile Range New Mexico 88002 Project Manager
Commcn Positioning & Navigation Systems Attn: Harold H. Bahr (AMCPM-NS-TM), Bldg 439 U.S. Army Electronics Command Fort Monmouth, New Jersey 07703 Director, Electronic Programs Attn: Code 427 Department of the Navy Washington, D.C. 20360 Commander
U.S. Naval Security Group Command Attn: G43
3801 Nebraska Avenue Washington, D.C. 20390
Naval Research Laboratory Washington, D.C. 20390 Attn: Code 2027 6 copies
Dr W.C. Hall, Code 7000 1 copy Dr A. Brodizinsky, Sup.Elec Div. 1 copy
Deputy Director and Chief Scientist Office of Naval Research Branch Office 1030 Est Gree Street
Pasadena, California 91101 Library (Code2124) Technical Report Section Naval Postgraduate School Monterey, California 93940 Glen A. Myers (Code 52Mv) Assoc Professor of Elec. Engineering Naval Postgraduate School Monterey, California 93940 Commanding Officer and Director U.S. Naval Underwater Sound Laboratory Fort Trumbull
New London, Connecticut 06840 Commanding Officer Naval Avionics Facility Indianapolis, Indiana 46241 Dr H. Harrison, Code RRE Chief, Electrophysies Branch National Aeronautics & Space Admin. Washington, D.C. 20546 NASA Lewis Research Center Attn: Library 21000 Brookpark Road Cleveland, Ohio 44135 Los Alamos Scientific Laboratory Attn: Report Library P.0. Box 1663 Los Alamos, New Mexico 87544 Federal Aviation Administration Attn: Admin Stds Div (MS-110) 800 Independence Ave S.W. Washington, D.C. 20590 Head, Technical Services Division Naval Investigative Service Headquarters 4420 North Fairfax Drive
Arlington, Virginia 22203 Commander
U.S. Naval Ordnance Laboratory Attn: Librarian
White Oak, Maryland 21502 (2 copies) Commanding Officer
Office of Naval Research Branch Office Box 39 FP0
New York, New York 09510 Conmanding Officer
Office of Naval Research Branch Office 219 South Dearborn Street Chicago, Illinois 60604 Commanding Officer
Office of Naval Research Branch Office 495 Summer Street
Boston, Massachusetts 02210 Dr G.M.R. Winkler
Director, Time Service Division U.S. Naval Observatory Washington, D.C. 20390
Commander (ADL) Naval Air Development Center Johnsville, Warminster, Pa 18974 Naval Air Systems Command
AIR 03
Washington, D.C. 20360 2 copies
Commanding Officer Naval Training Device Center Orlando, Florida 32813 Naval Ship Systems Conmand
Ship 031 Washington, D.C. 20360
Commander (Code 753) Naval Weapons Center Attn: Technical Library China Lake, California 93555 Naval ship Systems Conmand
Ship 035 Washington, D.C. 20360 U.S. Naval Weapons Laboratory Dahlgren, Virginia 22448
Commanding Officer Naval Weapons Center Corona Laboratories Attn: Library Corona, California 91720
Mr M. Zane Thornton, Chief Network Engineering, Communications and Operations Branch Lister Hill National Center for Biomedical Communications 8600 Rockville Pike Bethesda, Maryland 20014 U.S. Post Office Department Library - Room 1012 12th & Pennsylvania Ave, N.W. Washington, D.C. 20260 Director
Research Laboratory of Electronics Massachusetts Institute of Technology Cambridge, Massachusetts 02139 Mr Jerome Fox, Research Coordinator Polytechnic Institute of Brooklyn 55 Johnson Street
Brooklyn, New York 11201
Columbia Radiation Laboratory Columbia University 538 West 120th Street New York, New York 10027 Director
Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801 Director
Stanford Electronics Laboratories Stanford University Stanford, California 94305 Director
Microwave Physics Laboratory Stanford University Stanford, California 94305 Director, Electronics Research Laboratory University of California
Berkeley, California 94720 Director
Electronic Sciences Laboratory University of Southern California Los Angeles, California 90007 Director
Electronics Research Center The University of Texas at Austin Austin Texas 78712
Division of Engineering and Applied Physics 210 Pierce Hall
Harvard University Cambridge, Massachusetts 02138 Dr G.J. Murphy
The Technological Institute Northwestern University Evanston, Illinois 60201 Dr John C. Hancock, Head School of Electrical Engineering Purdue University Lafayette, Indiana 47907 Dept of Electrical Engineering Texas Technological College Lubbock, Texas 79409 Aerospace Corporation P.0. Box 95085
Los Angeles, California 90045 Attn: Library Acquisitions Group Proffessor Nicholas George California Inst of Technology Pasadena, California 91109 Aeronautics Library Graduat Aeronautical Laboratories California Institute of Technology 1201 E. California Blvd Pasadena, California 91109
Hie John Hop1ins University Applied Physics Laboratory Attn: Document Librarian 8621 Georgia Avenue Silver Spring, Maryland 20910 Raytheon Company Attn; Librarian Bedford, Massachusetts 01730 Raytheon Company Research Division Library 28 Seyon Street Waltham, Massachusetts 02154 Dr Sheldon J. Wells
Electronic Properties Information Center Mail Station E-175
Hughes Aircraft Company Culver City, California 90230
Yale University Engineering Department New Haven, Connecticut 06520 Airborne Instruments Laboratory Deerpark, New York 11729 Raytheon Company Attn: Librarian Bedford, Massachusetts 01730 Lincoln Laboratory
Massachusetts Institute of Technology Lexington, Massachusetts 02173 The University of Iowa The University Libraries Iowa City, Iowa 52240 Dr Robert E. Fontana
Systems Research Laboratories Inc. 7001 Indian Ripple Road Dayton, Ohio 45440
Lenkurt Electric Co,Inc 1105 County Road San Carlos, California 94070 Attn: Mr E.K. Peterson Nuclear Instrumentation Group
Bldg 29, Room 101 Lawrence Radiation Laboratory University of California Berkeley, California 94720
Philco Ford Corporation Communications & Electronics Div. Union Meeting and Jolly Rods Blue Bell, Pennsylvania 19422 Sylvania Electronic Systems
Applied Research Laboratory Attn: Documents Librarian 40 Sylvan Road Waltham, Massachusetts 02154 Hollander Associates P.0. Box 2276 Fullerton, California 92633
Union Carbide Corporation Electronic Division P.0. Box 1209
Mountain View, California 94041 Electromagnetic Compatibility Analysis Center (ECAC), Attn: ACLP
North Severn Annapolis, Maryland 21402 Illinois Institute of Technology
Dept of Electrical Engineering Chicago, Illinois 60616
Director
U. S. Army Advanced Materiel Concepts Agency
ADDENDUM15
The University of Arizona Dept of Electrical Engineering Tucson, Arizona 85721 Utah State University Dept Of Electrical Engineering Logan, Utah 84321
Dept of Electrical Engineering Rice University
Houston3 Texas 77001
Case Institute of Technology Engineering Division University Circle Cleveland, Ohio 44106 Hunt Library Carnegie-Melion University Schenley Park Pittsburgh, Pennsylvania 15213 Dr Leo Youns
Stanford Research Institute Menlo Park, California 94025 School of Engineering Sciences Arizona State University Tempe, Arizona 85281
Engineering & Mathmatical Sciences Library University of California at Los Angeles 405 Hilgred Avenue
Los Angeles, California 90024 The Library
Government Publications Section University of California Santa Barbara, California 93106 Carnegie Institute of Technology Electrical Engineering Department Pittsburgh, Pennsylvania 15213 Professor Joseph E. Rowe
Chairman, Dept of Electrical Engineering The University of Michigan Ann Arbor, Michigan 48104 New York University College of Engineering New York, New York 10019
ERRATUM
Syracuse University Dept of Electrical Engineering Syracuse, New York 13210
Mr Jerome Fox, Research Coordinator Polytechnic Institute of Brooklyn 55 Johnson St (Should be 333 Jay St) Brooklyn, N.Y. 11201
DELETE
Mr Morton M. Pavane, Chief
AFSC Scientific & Tech. Liaison Office 26 Federal Plaza, Suite 1313
New York, N. Y. 10007 Commanding Officer
Office of Naval Research Branch Office Box 39 FPO
New York, N. Y. 09510
Research Laboratories for the Eng. Sc. School of Engineering & Applied Science University of Virginia
Charlottesville, Virginia 22903 Dept of Electrical Engineering College of Engineering & Technology Ohio University
Athens, Ohio 45701 Project Mac
Document Room
Massachusetts Institute of Technology 545 Technology Square
Cambridge, Massachusetts 02139 Lehigh University
Dept of Electrical Engineering Bethelhelm, Pennsylvania 18015 Commander Test Command (TCD-) Defense Atomic Support Agency Sandia Base
Albuquerque, New Mexico 87115
Materials Center Reading Room 13-2137 Massachusetts Institute of Technology Cambridge, Massachusetts 02139
Professor James A. Cadzow
Department of Electrical Engineering State University of New York at Buffalo Buffalo, New York 14214
Director, Naval Research Laboratory Attn: Library, Code 2029 (ONRL) Washington, D.C. 20390
Commanding Officer (Code 2064) Navy Underwater Sound Laboratory Fort Trumbull
Security Classification^
DOCUMENT CONTROL DATA - R & D ( S e c u r i t y c l a s s i f i c a t i o n o f t i t l e , b o d y o r a b s t r a c t a n d i n d e x i n g a n n o t a t i o n m u s t
I O R I G I N A T I N G A C T I V I T Y ( C o r p o r a t e a u t h o r )
University of Illinois
Coordinated Science Laboratory Urbana, Illinois 61801
3 . R E P O R T T I T L E
be e n t e r e d w h e n t h e o v e r a l l r e p o r t i s c l a s s i f i e d )
\ z a . R E P O R T S E C U R I T Y C L A S S I F I C A T I O N 2 b . G R O U P
AN ITERATIVE APPROACH FOR THE CORRECTION OF ITERATIVE ERRORS
4 . d e s c r i p t i v e N O T E S ( T y p e o f r e p o r t a n d i n c l u s i v e d a t e s )
5. a u t h o r( S ) ( F i r s t n a m e , m i d d l e i n i t i a l , l a s t n a m e )
CHIEN, Robert T. & HONG, Se June
6 R E P O R T D A T E November, 1969 7 a . T O T A L N O . O P P A G E S 13 7 b . N O . O F R E F S
12
8 a . C O N T R A C T O R G R A N T N O . 9 a . O R I G I N A T O R ’ S R E P O R T N U M B E R ( S )DAAB-07-67-C-0199; also NSF Grant Gk-2339
b . P R O J E C T N O .
R-443
9 b . O T H E R R E P O R T N O ( S ) ( A n y o t h e r n u m b e r s t h a t m a y b e a s s i g n e d t h i s r e p o r t )
1 0 . D I S T R I B U T I O N S T A T E M E N T
This document has been approved for public release and sale; its distribution is unlimited«,
11 S U P P L E M E N T A R Y N O T E S 1 2 . S P O N S O R I N G M I L I T A R Y A C T I V I T Y
Joint Services Electronics Program thru Uo S. Army Electronics Command Fort Monmouth, New Jersey 07703____
1 3 . A B S T R A C T
The errors most likely to occur in a high-speed multiplier are called the iterative errors» An arithmetic coding technique for correction of such error patterns is proposed. We present a class of codes and show its error correcting ability» The unique feature of this code is an iterative decoding method.