---
title: 'Sum-Rank Metric Codes: Theory & Applications'
url: https://www.emergentmind.com/topics/sum-rank-metric-codes
type: topic
---

# Sum-Rank Metric Codes: Theory & Applications

Sum-rank metric codes are codes on block-structured ambient spaces in which the distance between two codewords is the sum of the ranks of the blockwise differences. In matrix form, this places codewords in products of matrix spaces; in vector form, it places them in block decompositions of \(\mathbb F_{q^m}^n\). The resulting metric interpolates exactly between the rank metric and the Hamming metric, and it is used in settings such as multishot network coding, distributed storage, and space-time coding. Recent work has developed a substantial theory around these codes, including support and anticode formalisms, MSRD extremality, asymptotic bounds, generic and list decoding, explicit algebraic and combinatorial constructions, and equivalence invariants [1808.02468] [2307.13613] [2001.04812] [2503.10234] [2507.04158].

## 1. Formal models and basic metric structure

A standard matrix model fixes blocks \(X_i\in \mathbb F_q^{m\times \eta}\) and works in
\[
M^\ell=\prod_{i=1}^{\ell}\mathbb F_q^{m\times \eta},
\qquad
\boldsymbol X=(X_1,\dots,X_\ell).
\]
The sum-rank weight is
\[
\operatorname{wt}_{\mathrm{srk},\ell}(\boldsymbol X)=\sum_{i=1}^{\ell}\operatorname{rank}(X_i),
\]
and the induced distance is
\[
d_{\mathrm{srk},\ell}(\boldsymbol X,\boldsymbol Y)=\sum_{i=1}^{\ell}\operatorname{rank}(X_i-Y_i).
\]
Under the common assumption \(\eta\le m\), the total code length is \(n=\eta\ell\), the maximum weight is \(n\), and the metric is translation-invariant and invariant under multiplication by nonzero field scalars. In this setting, a general sum-rank code is any nonempty subset \(C\subseteq M^\ell\), while an \(\mathbb F_q\)-linear sum-rank code is a subspace \(C\le M^\ell\) [2503.10234].

A complementary vector model writes \(x\in \mathbb F_{q^m}^n\) in \(\ell\) blocks of length \(\eta=n/\ell\),
\[
x=[x_1\mid \cdots \mid x_\ell],\qquad x_i\in \mathbb F_{q^m}^{\eta},
\]
and expands each block over \(\mathbb F_q\). The block rank weight is the \(\mathbb F_q\)-rank of the expansion matrix, and the sum-rank weight is the sum of those block ranks. In this notation, the weight decomposition \([t_1,\dots,t_\ell]\) records the rank of each block, with \(\sum_i t_i=\operatorname{wt}_{SR}(x)\). This model is especially convenient for decoding and asymptotic analysis over extension fields [2001.04812] [2102.02244].

The metric contains the two classical extremes exactly. If \(\ell=1\), there is a single block and the sum-rank metric is the ordinary rank metric. If every block is \(1\times 1\), or equivalently every vector block has length \(1\), then each nonzero block contributes rank \(1\), and the sum-rank metric becomes the Hamming metric. More general formulations also allow unequal block sizes
\[
\mathcal M=\bigoplus_{i=1}^t \mathbb F_q^{\,n_i\times m_i},
\]
with
\[
\operatorname{srk}(X)=\sum_{i=1}^t \operatorname{rk}(X_i),
\]
which makes the hybrid Hamming/rank nature explicit at the level of the ambient geometry [2307.13613].

## 2. Supports, anticodes, generalized weights, and MSRD structure

A foundational development is the support theory of sum-rank codes. For a codeword \(\mathbf c=(\mathbf c^{(1)},\dots,\mathbf c^{(\ell)})\), the sum-rank support is the tuple of row spaces of the block expansion matrices over the corresponding base fields. These supports form a lattice
\[
\mathcal P(K^{\mathbf n})=\mathcal P(K_1^{n_1})\times\cdots\times \mathcal P(K_\ell^{n_\ell}),
\]
with componentwise sum, intersection, and orthogonal complement. The associated support spaces
\[
\mathcal V_{\mathcal L}=\{\mathbf c\in \mathbb F^n:\operatorname{Supp}(\mathbf c)\subseteq \mathcal L\}
\]
are in lattice isomorphism with supports, satisfy
\[
\dim(\mathcal V_{\mathcal L})=\operatorname{Rk}(\mathcal L),
\qquad
\mathcal V_{\mathcal L}^\perp=\mathcal V_{\mathcal L^\perp},
\]
and provide the natural framework for restriction, shortening, duality, and effective length. In particular, the generalized sum-rank weights \(d_{SR,r}\) admit Wei-type duality, and degeneracy is characterized by \(d_{SR}(\mathcal C^\perp)=1\) [1808.02468].

A second structural axis is the anticode viewpoint. For a linear code \(\mathcal C\subseteq \mathcal M\subseteq \prod_i \mathbb F_q^{m_i\times n_i}\), the sum-rank Anticode Bound states
\[
\dim(\mathcal C)\le \max_{C\in \mathcal C}\left\{\sum_{i=1}^{\ell} m_i\,\operatorname{rk}(C_i)\right\}.
\]
Codes attaining equality are optimal anticodes. These are classified: an optimal sum-rank anticode is precisely a product of optimal rank-metric anticodes on the blocks with \(m_i>1\), together with an optimal Hamming-metric anticode on the \(1\)-row part. This classification is then used to define generalized sum-rank weights through intersections with optimal anticodes, extending both generalized Hamming weights and generalized rank weights [2107.13611].

The same anticode framework clarifies MSRD codes. In the equal-row case, the Singleton-type bound becomes the usual
\[
d\le n-k+1
\]
in the vector model, while in the general matrix/block model it is expressed through the decomposition of \(d-1\) across block lengths. Codes meeting the corresponding bound are MSRD. Their generalized weights are parameter-determined, in direct analogy with MDS and MRD codes. The appendix of the anticode paper also shows that the support-space version of generalized weights measures information leakage in multishot network coding, linking the structural theory to secure communication [2107.13611].

## 3. Upper bounds, asymptotics, and covering theory

The counting theory of sum-rank spheres and balls is central to asymptotic bounds. In the equal-block vector model, the number of words of weight \(t\) is a sum over bounded ordered partitions \((t_1,\dots,t_\ell)\), and the dominant exponent of a sphere of total weight \(t\) is
\[
(m+\eta)t-\frac{t^2}{\ell}.
\]
This leads to simplified sphere-packing and Gilbert–Varshamov bounds for linear sum-rank codes, together with asymptotic tradeoffs in both the classical regime of fixed block size and growing number of blocks, and the newer regime in which block size itself grows with code length. In that latter regime, with \(m=\eta\xi\), the asymptotic GV-like rate behaves as
\[
\mathcal R \sim 1-\delta\left(1+\frac1\xi\right)+\frac{\delta^2}{\xi},
\]
and random linear sum-rank codes are shown to attain near-GV behavior with high probability; for sufficiently large extension fields, random linear codes are also MSRD with high probability [2102.02244].

For finite-parameter upper bounds, graph methods have become important. The sum-rank-metric graph has vertex set \(\mathcal M\), adjacency given by sum-rank distance \(1\), and decomposes as a Cartesian product of bilinear forms graphs. This graph is vertex-transitive and walk-regular, but it is distance-regular only in the Hamming-like case where all blocks are \(1\times m\) of the same width. Ratio-type spectral bounds applied to this graph yield new explicit upper bounds for minimum distances \(d=3,4\), improve earlier bounds in many parameter sets, and produce new non-existence results for possibly nonlinear MSRD codes [2307.13613].

A complementary development is a genuine Delsarte-style linear programming bound. Although the natural sum-rank distance relations do not generally form an association scheme directly, the Cartesian product decomposition embeds the problem into a direct product of bilinear forms schemes. The resulting LP imposes zero inner-distribution entries on all relation classes whose blockwise rank distances sum to less than the target minimum distance, and computational experiments show that the bound outperforms all previously known bounds on a range of relatively small instances [2406.15926].

Covering theory adds a different extremal perspective. For prescribed sum-rank covering radius \(\rho\), the minimum cardinality of a covering code is denoted \(K(\mathbb F_{q^m}^n,\rho)\). The basic sphere-covering bound is
\[
K(\mathbb F_{q^m}^n,\rho)\ge \frac{q^{mn}}{\operatorname{Vol}_{SR}^{\le}(\rho)},
\]
and a nontrivial MSRD-based upper bound is
\[
K(\mathbb F_{q^m}^n,\rho)\le q^{(m-\lfloor \rho/\ell\rfloor)(n-\rho)}.
\]
The same paper adapts elementary linear subspaces to the sum-rank setting and identifies ball-intersection formulas as the main missing ingredient for sharper recursive lower bounds [2210.02282]. A later covering-code transfer from Hamming metric to sum-rank metric yields stronger Singleton-like bounds when block lengths are large, and also leads to explicit quasi-perfect and distance-optimal families, especially for minimum sum-rank distance \(4\) [2601.05581].

## 4. Decoding, hardness, and list decoding

Generic decoding in the sum-rank metric was developed through support guessing. For an \(\mathbb F_{q^m}\)-linear code with parity-check matrix \(H\), the error model combines a blockwise weight decomposition with row and column support notions: each block error \(e_i\) admits a factorization \(e_i=a_iB_i\), which yields a row support as the product of the row spaces of the \(B_i\), and a column support as the product of the column spaces of the expansions of the \(a_i\). Knowing a row super-support or column super-support reduces decoding to a linear system. The paper gives erasure-decoding complexities \(O((n-k)^3m^2)\) and \(O((n-k)^3m^3)\) operations over \(\mathbb F_q\), respectively, and then builds a Las Vegas generic decoder that repeatedly samples random supports. Its expected work factor is controlled by a quantity \(Q_{\zeta,\ell}(t,s)\), and a simple upper bound has exponential term
\[
q^{\,t\left(\zeta-\frac{s}{\ell}\right)}.
\]
The same work also proves a probabilistic hardness reduction from decisional Hamming syndrome decoding to decisional sum-rank syndrome decoding; under the stated extension-degree condition, \(D\ell\)-SRSD in \(ZPP\) would imply \(NP=ZPP\) [2001.04812].

For highly structured small-block codes, faster decoding is possible. Binary linear \(2\times 2\) sum-rank-metric codes built from quaternary BCH, Goppa, and additive quaternary codes admit a reduction from sum-rank decoding to Hamming decoding over \(\mathbf F_4\). In the main decoding theorem, one decodes once in one constituent code and three times in twists of the other constituent, and this yields quadratic-time decoders with cost \(O(t^2)\) operations in \(\mathbf F_4\). The paper also constructs asymptotically good sequences of quadratically encodable and decodable binary \(2\times 2\) sum-rank codes satisfying
\[
R_{sr}(\delta_{sr}) \ge 1-\frac{1}{2}\left(H_4\!\left(\frac{4}{3}\delta_{sr}\right)+H_4(2\delta_{sr})\right)
\]
from Goppa codes [2311.03619].

List decoding has recently reached a capacity theorem for the sum-rank metric. In the matrix model \(M^\ell=\prod_i \mathbb F_q^{m\times \eta}\), with \(b=\eta/m\in(0,1]\), the asymptotic log-volume exponent of a radius-\(\rho n\) ball is
\[
\kappa_b(\rho)=\rho+\rho b-\rho^2 b,
\]
so the capacity expression is
\[
1-\kappa_b(\rho)=1-(\rho+\rho b-\rho^2 b).
\]
Random general sum-rank codes of rate \(1-\kappa_b(\rho)-\epsilon\) are \((\rho,O(1/\epsilon))\)-list-decodable with high probability, and the same optimal \(O(1/\epsilon)\) list size is proved for random \(\mathbb F_q\)-linear sum-rank codes after establishing a limited correlation property between sum-rank balls and \(\mathbb F_q\)-subspaces. A notable nuance is that the specialization \(m=\eta=1\) yields \(1-\rho(2-\rho)\), which does not recover the classical Hamming list-decoding capacity \(1-H_q(\rho)\) [2503.10234].

## 5. Explicit constructions and geometric families

A geometric construction of sum-rank codes is given by linearized Algebraic Geometry codes. Starting from a cyclic Galois cover of curves \(Y\to X\), the construction uses Ore polynomial quotients
\[
D_{L,x}=L[T;\Phi]/(T^r-x),
\]
noncommutative Riemann–Roch spaces \(\Lambda_{L,x}(E)\), and evaluations into products of endomorphism algebras
\[
End_k(V_1)\times \cdots \times End_k(V_s).
\]
The resulting codes satisfy a Goppa-like distance bound
\[
d\ge sr-\deg_Y(E),
\]
together with a lower bound on the dimension derived from the noncommutative Riemann–Roch theorem. In genus \(0\), the construction recovers linearized Reed–Solomon codes exactly, while over asymptotically optimal towers of curves it yields families asymptotically better than the sum-rank Gilbert–Varshamov bound [2303.08903].

A more explicit \(2\times 2\) algebraic-geometric route uses quadratic Kummer extensions. Two constructions, one built from a totally ramified place and one from a completely split place, produce \(2\times 2\) sum-rank codes with parameters
\[
[4s,\ l_k,\ d_{sr}\ge 2s-k].
\]
Their determinant analysis relies on local integral bases and valuation control, and the paper emphasizes that these constructions can have superior code lengths compared with linearized Reed–Solomon codes under equivalent parameter constraints. An elliptic-function-field example gives concrete codes of parameters \([24,6,6]\) [2506.18653].

Several combinatorial constructions target small fields and special weights. Binary \(2\times 2\) codes obtained from BCH, Goppa, and additive quaternary codes can exceed the dimensions of sum-rank BCH codes with the same guaranteed minimum distance, and the exact weight formula
\[
\operatorname{wt}_{sr}(a_2x+a_1x^2)
=2\operatorname{wt}_H(a_1)+2\operatorname{wt}_H(a_2)-3|I|
\]
explains the refinement over earlier lower bounds [2311.03619]. A geometric study of one-weight codes identifies \(k\)-dimensional \(\mathbb F_{q^m}\)-linear sum-rank codes with tuples of \(\mathbb F_q\)-subspaces of \(\mathbb F_{q^m}^k\), proves that doubly-extended linearized Reed–Solomon codes remain MSRD, and shows that when \(k=2\) they are one-weight. The same work introduces \(n\)-simplex codes in the sum-rank metric via Singer subgroup orbits and constructs a new family of one-weight MSRD codes for \(q=2\), called \(2\)-fold linearized Reed–Solomon codes [2112.04989].

Distance-optimal and quasi-perfect families have also been constructed from covering-code ideas. The covering lift from Hamming metric produces infinite families of quasi-perfect \(q\)-ary sum-rank codes with matrix sizes \(2\times m\), explicit binary quasi-perfect codes of matrix size \(2\times 2\), distance-optimal sum-rank codes of matrix sizes \(s\times s\) and \(2\times 2\) with minimum distance \(4\), and almost-MSRD \(2\times 2\) codes of block length up to \(q^2\) and Singleton defect \(2\) [2601.05581].

## 6. Geometry of equivalence, invariants, and current directions

The code equivalence problem in the sum-rank metric has motivated a new invariant theory. In the matrix model \(\Mat(\mathbf m,\mathbf n,\mathbb F_q)\), additive isometries are classified as block permutations followed by blockwise rank-metric isometries. Building on this, the invariant paper introduces \(2^t\) generalized idealisers
\[
\mathcal I_{\mathbf s}(\mathcal C),
\]
the centraliser \(\mathrm C(\mathcal C)\), the center \(Z(\mathcal C)\), and a refined notion of \(\mathbb F_{q^i}\)-linearity defined via the left idealiser rather than the ambient coordinate model. The associated nuclear parameters
\[
\bigl(|\mathcal C|,\ |I_\ell(\mathcal C)|,\ |I_r(\mathcal C)|,\ |\mathrm C(\mathcal C)|,\ |Z(\mathcal C)|\bigr)
\]
are equivalence invariants. A skew-polynomial realization \(R/RH_\Lambda\) makes these quantities explicitly computable, and the paper uses them to separate the major known MSRD families, including linearized Reed–Solomon, additive twisted linearized Reed–Solomon, and TZ-type twisted linearized Reed–Solomon codes [2507.04158].

Several open directions are stated explicitly across the literature. The list-decoding capacity work is existential and does not provide explicit constructions or efficient list-decoding algorithms at the capacity radius; it also identifies random \(\mathbb F_{q^m}\)-linear vector sum-rank codes as an unresolved case [2503.10234]. The covering paper identifies explicit formulas for ball intersections \(\operatorname{Vol}(\rho,\delta)\) as the main missing piece for stronger lower bounds [2210.02282]. The algebraic-geometric papers point to the construction of MSRD algebraic-geometric sum-rank codes as a natural next step [2303.08903] [2506.18653]. The covering-code approach to strong Singleton-like bounds leaves open the construction of almost-MSRD families with longer block lengths and larger minimum distance [2601.05581]. Taken together, these directions show that the field has moved beyond the initial definition of the metric toward a mature research program in asymptotic theory, extremal combinatorics, algorithmics, and equivalence classification.

Source: https://www.emergentmind.com/topics/sum-rank-metric-codes