432 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 41, NO.2. MARCH 195 Bounds on the Minimum Support Weights Tor Helleseth, Member IEEE, Torleiv Kl)ve, Senior Member, IEEE, Vladimir . Levenshtein, Member IEEE, and 0yvind Ytrehus, Member EEE - The minimum support weight, d (C). of a linear code C over GF(q) is the minimal size of the support of an r-dimensional subcode of C.  number of bounds on r () are derived, generaliing the Plotkin bound and the Griesmer bound, as well as giving two new eistenial bounds.  the main result, it is shown that there exist codes of any iven rate R whose ratio d/d is lower bounded by a number raning from ( q r )/(" _ q.- l ) to , depending on R.  Terms-Linear codes, support weight. bounds. . INTRODUCTION F OR any code D, X(D), the support of , is the set of positions where not all the codewords of D are zero, and w s(D). the support weight of D, is the size of X()V For a linear eode C of lengh n and dimension k over GF(q') (for short: an [n, kq] code) and any integer T, where 1 T � k, the rth minimum support weight is deined by dT = dT(C) = min{ws(D ) I D is an [n,T; q] subcode of C. In particular, the minimum distance of C is clI. The weight hierarchy of C is the set {lI, d,"',dW The rth minimum support weight is also known as the Tth generalized Ham ming weight [35]. The weight hierarchy has been called the length/dimension poile [7]. The weight hierarchy has been studied by a number of researchers in the last few years. In the Reference secion we have listed all papers, known to us, that deal with support weights and related questions.  mention here one lemma that will be used later . A proof for the binary case was given in [14], the general ease ꜳ be proved similarly. Let Fk,l be the set of l-dimensional subspaces of GF(q) k. Let G be a ixed k x n generator matrix for an [n, k; q] code C. For each   GF (q) k , let m(x) be the number of times  appears as a column in G. For each subset  of GF(q ) k, let m(U) = L m(x). E Lemma 1 For 1 � r � k we have dT(C) = n-  {m(U) I U E F k , k -r} = min {m(GF(q) k \ U) I U  Fk,k-r}. Manuscript received November 24, 1993; revised July 15, 1994. This research was supported by The Norwegian Research Council under Grant 463.93/005. T. HelleseLh, T. Kl0ve, and 0. Ylrehus are with the Department of Informatics. Cniversily of Bergen, HlB, N-5020 Bergen, Nor way. . L Levenshtein is with the Keldysh Instilute for Applied Mathematics, Russian Academy of Sciences. 125047 Moscow, Russia. IEEE Log Number 9408364. In this paper we give a number of bounds which generalize known bounds involving n, k, and d. A basic tool for the proof of the first two is the fundamental relation (- q)d r 2 (q" - 1)r1r-1. This was proved in [14] for q = 2, the proof given there generalizes directly to general q. Below we will give a diferent proof using a method which is of interest i n its own right. In the last and main part of the paper we give existence results. II.  FUNDAMENTAL RELATION Lemma 2 For any [n, q] eode D we have L w(') = qT-I(q - l)ws(D). c'D Proof' In each position in XD), exactly q" _ "- codewords in  are nonzero in that position. • For an [n, rq] code D and any 8 sueh that 0 8 r, let V( D, 8 ) be the set of subspaces of D of dimension 8, the number of such subspaees i s given by the Gaussian binomial coeicients. Since wee) = ws({ae I  E G(q)) the following is a generalization of Lemma 2. ea 3 For any [n, rq] code D ad any 8 such that 8  T we have " [1'-1]  ws( V ) = q r -s 8 _ 1 ws(D). VEV(D,s)  Proof: By Lemma 2 we have 1 L ws (V ) = L 8-1 1 LW ( c ) VEV(D,8) VEV(D,8) q (q - ) cEV 1 q S-l (q _ 1) L _ w e e ) L 1 cED\{O} VEV(D,s) S.t. cEV 1 " [r-l] = qS_l ( q_l) w ( e ) 8 -1 q = q S - ( _ 1) q r -I (q - l) ws(D ) [: : ] q = q l'-S [: : ]  ws(D). 0018-9448/95$04.00  1995 IEEE