--- title: Routing Codes in Quantum and Network Applications url: https://www.emergentmind.com/topics/routing-codes type: topic --- # Routing Codes in Quantum and Network Applications Routing codes is a polysemous technical term. In the 2026 quantum-error-correction literature, it denotes a family of quantum low-density parity-check codes defined to combine high rate and threshold with short, mutually parallel non-local connectivity, explicitly targeting implementability on superconducting and neutral-atom hardware [2606.25330]. Earlier communication-theoretic literature also used the phrase for coding schemes designed around routed networks rather than fully mixed network coding, including joint multi-path routing with adaptive FEC, subset codes for packet-switched routing, subspace codes in simplified routing-based interference models, and constant-weight lattice codes for store-and-forward routing [0611086][1210.7341][2304.13500][1301.7345]. ## 1. Quantum routing codes as a qLDPC family Routing codes are defined on an even–even torus $$ L=\mathbb Z_l\times\mathbb Z_m,\qquad l,m\ \text{even}, $$ with group ring $$ R=\mathbb F_2[x,y]/(x^l-1,\ y^m-1). $$ Each lattice site \((i,j)\in L\) corresponds to the monomial \(x^i y^j\). Qubits are partitioned by the parity of \(i+j\) into data qubits $$ D=\{(i,j)\mid i+j\ \text{even}\}, $$ \(X\)-syndrome qubits $$ X=\{(i,j)\mid i\ \text{odd},\ j\ \text{even}\}, $$ and \(Z\)-syndrome qubits $$ Z=\{(i,j)\mid i\ \text{even},\ j\ \text{odd}\}. $$ The construction chooses routing vectors \(\{v_1,\dots,v_T\}\) for \(X\)-syndromes and \(\{w_1,\dots,w_T\}\) for \(Z\)-syndromes; at time step \(t\), each \(X\)-syndrome at \(r\) swaps via iSWAP with data at \(r+v_t\), and each \(Z\)-syndrome at \(r\) swaps with data at \(r+w_t\) [2606.25330]. The associated involutive permutation is $$ \sigma_t(u)= \begin{cases} u+v_t,& u\in X,\\ u+w_t,& u\in Z,\\ u-v_t,& u\in D\ \text{and}\ u-v_t\in X,\\ u-w_t,& u\in D\ \text{and}\ u-w_t\in Z. \end{cases} $$ With \(S_t=\sigma_1\circ\cdots\circ\sigma_t\), an \(X\)-syndrome initially at \(x\) collects parity from $$ q_{x,t}=S_{t-1}^{-1}\circ \sigma_t\circ S_{t-1}(x), $$ and similarly for \(Z\)-syndromes. Fixing reference syndrome positions \(x\in X\) and \(z\in Z\), the stabilizer supports are encoded by $$ P(x,y)=\sum_{(a,b)\in Q_x}x^a y^b,\qquad Q(x,y)=\sum_{(c,d)\in Q_z}x^c y^d. $$ All \(X\) stabilizers are generated by even–even translations of \(P\), and all \(Z\) stabilizers by translations of \(Q\). In CSS form, $$ H=[H_X\mid H_Z]. $$ Commutativity is characterized by Proposition 1: the stabilizers commute iff, in $$ P(x,y)\cdot Q(x^{-1},y^{-1})=\sum_{i,j} c_{i,j}x^i y^j, $$ every term \(x^i y^j\) with \(i,j\) both even has \(c_{i,j}=0\) in \(\mathbb F_2\). Proposition 2 gives a sufficient symmetry condition: if \(v_t=w_t\) for all \(t\) and \(v_t=v_{T-t+1}\), then commutativity is automatic. The explicit constructions in the report use a time-reversal symmetric sequence of length \(T=7\) for weight-7 codes, containing one non-local vector plus local steps \((1,0)\) and \((0,1)\). The logical dimension satisfies $$ k=n-\operatorname{rank}(H_X)-\operatorname{rank}(H_Z),\qquad n=lm, $$ and \(d\) is computed via integer-programming on \(H_X,H_Z\). ## 2. Connectivity reduction and parallel non-locality The distinctive hardware feature of routing codes is not merely the presence of non-local couplings, but the simultaneous reduction of coupling length, qubit connectivity, and routing conflict. Each qubit has degree $$ \deg=D_{\text{local}}+D_{\text{nonlocal}}, $$ where \(D_{\text{local}}=2\) or \(3\) and \(D_{\text{nonlocal}}\) equals the number of distinct non-local vectors, which is \(1\) in the explicit constructions. For the \([[70,8,7]]\) instance, the non-local vector is \(v=(2,1)\), with $$ \ell_{\max}=\|v\|_2=\sqrt5\approx 2.24. $$ The comparison given in the report is with bivariate bicycle codes using \(\{(6,3),(3,6)\}\), giving \(\ell_{\max}\approx \sqrt{45}\approx 6.7\); this reduces maximum coupling length by \(\gtrsim 3\times\) [2606.25330]. For weight-7 routing codes, the check weight is \(w=T=7\), identical to BB codes, but qubit connectivity is only \(4\) or \(5\) instead of \(6\). Proposition 3 states that, on a sufficiently large torus that avoids path wrapping, time-reversal symmetric and non-negative \(\{v_t\}\) guarantee that no two CXSWAP gates conflict temporally or spatially, so all gates assigned to a given step \(t\) can be implemented in parallel. Because all non-local edges share the same direction \(v^\*\), no wiring crossings occur in a multi-layer superconducting layout and AOD-shuttled atom trajectories never intersect in neutral-atom arrays. This combination of short-range non-locality and mutual parallelism is the central differentiator of the family. A plausible implication is that the usual trade-off between high-rate qLDPC structure and hardware routability is being shifted from a question of whether non-locality exists to a question of how structured that non-locality is. ## 3. Parameters, thresholds, and overhead The report lists selected instances \[ [[54,8,6]],\quad [[70,8,7]],\quad [[100,8,10]], \] with rates \[ R=8/54\approx 0.148,\qquad R\approx 0.114,\qquad R=8/100=0.08, \] and distances \(d=6,7,10\), respectively. In these families \(k\) remains fixed at \(8\) as \(n\) grows, so \(R\sim O(1/n)\); the report also notes that one may increase \(k\) by tiling multiple \(X/Z\) orbits, although the focus is on fixed-\(k\) instances [2606.25330]. Under circuit-level simulation with Stim for depolarizing noise, using a two-qubit error rate \(p\) after each iSWAP and depolarized single-qubit idles, and BP-OSD decoding for all codes, the reported threshold is approximately \(0.5\%\). For \(p<10^{-3}\), the \([[140,8,12]]\) routing code achieves \(p_L<10^{-12}\). Relative to the rotated surface code with \(d=6\)–\(12\) at \(p=10^{-3}\), routing codes require \(\sim 8\times\) fewer physical qubits for the same logical error rate. The circuit-level distance \(d_{\mathrm{circ}}\) equals the theoretical \(d\) or exceeds the BB codes’ \(d_{\mathrm{circ}}\) on instances with equal \(d\). | System | Connectivity and \(\ell_{\max}\) | Rate / hardware score | |---|---|---| | Routing \([[70,8,7]]\) | \(\deg=3+1\), \(\ell_{\max}=\sqrt5\approx2.24\) | \(R\approx0.114\), \(C_{\mathrm{hw}}\approx1.64\) | | BB \([[72,12,6]]\) | \(\deg=4+2\), \(\ell_{\max}\approx6.7\) | \(R=0.167\), \(C_{\mathrm{hw}}\approx2.01\) | | Surface code \(d=7\) | \(\deg=4\) local, \(\ell_{\max}=1\) | \(R\approx1/d^2\approx0.02\), \(C_{\mathrm{hw}}=1\) | The comparison given in the report summarizes the design point succinctly: routing-code rates \(0.08\)–\(0.15\), BB rates \(0.074\)–\(0.167\), surface-code rates \(\ll 0.1\); routing-code connectivity \(4\)–\(5\), BB connectivity \(6\), surface connectivity \(4\); routing-code \(\ell_{\max}\sim2\)–\(4\), BB \(\sim6.7\), surface \(1\); and qubit overhead reduction \(\gtrsim 8\times\) relative to surface codes. ## 4. Hardware mapping in superconducting and neutral-atom platforms The report provides platform-specific implementation guidance. In superconducting multi-layer layouts, using the HAL framework, routing codes require \(3\)–\(4\) routing tiers versus \(5\)–\(6\) for BB codes; average coupler length is approximately \(5\)–\(11\) unit-spacings versus approximately \(9\)–\(15\); average bump bonds per edge are \(\sim 3\) versus \(\sim 4\)–\(5\); TSVs per edge are \(\sim 2.1\) versus \(\sim 3\); and the composite hardware-complexity score is \(C_{\mathrm{hw}}=1.57\)–\(1.96\) versus \(2.01\)–\(2.24\). The implementation recommendation is to place each non-local vector on a single higher tier and exploit mutual parallelism to avoid via crossings. For neutral-atom AOD scheduling, the shuttle distance for \((2,1)\) is \(\sqrt5\approx2.24\), compared with \(\sqrt{45}\approx6.7\) for the BB vector \((6,3)\); conflict-graph coloring yields \(2\) time steps for routing codes versus up to \(4\) for BB codes, and the suggested schedule is a symmetric forward-and-backward routing cycle [2606.25330]. These claims sit within a broader hardware-aware qLDPC-routing landscape. HAL, introduced for arbitrary non-local QECC placement and routing in multi-layer superconducting hardware, defines raw layout metrics \(Q_1\) through \(Q_4\) and combines them into $$ C_{\rm hw}=1+\frac{\sum_{i=1}^4 w_i c_i}{\sum_{i=1}^4 w_i}, $$ with the rotated surface code as the \(C_{\rm hw}=1\) baseline; it was used to compare BB, tile, and radial-code families and to quantify multi-tier cost directly [2507.23011]. A different hardware-software co-design for BB codes uses a programmable 2D toric oscillator network, reducing long-range couplers from \(O(n)\) to \(O(\sqrt n)\), with symmetry-aware routing and fully parallel long-range gates within each syndrome-extraction pass [2604.18714]. A common misconception is that hardware-aware qLDPC implementation is exhausted by finding any long-range routing substrate. The comparison between routing codes, HAL-style placement studies, and programmable toric fabrics indicates a narrower issue: the geometry, directionality, and scheduling structure of non-local interactions materially affect tiers, crossings, cycle duration, and transport conflicts. ## 5. Earlier communication-theoretic meanings Before the qLDPC usage, “routing codes” appeared in networked communication settings where coding was designed explicitly for routed traffic. In real-time streaming, the term denoted the joint design of multi-path routing and adaptive FEC. The key scalar was the Redundancy Overall Requirement, $$ \mathrm{ROR}=\sum_{e\in E} p_e\Bigl(\frac{N(\alpha_e)}{M}-1\Bigr), $$ which rates a multi-path routing topology by the total adaptive redundancy required under link failures. The capillary routing algorithm constructed multi-path patterns and the reported trend was that further diversity, when achieved by capillary routing, reduced the overall requirement in FEC codes rather than increasing it [0611086]. For routed packet networks, subset codes provided a coding-theoretic framework in which a transmission is a subset \(X\subseteq S\) of packets and the receiver sees \(Y\subseteq S\) under permutation, deletions, insertions, and errors. The defining metric is the symmetric-difference distance $$ d(X,Y)=|X\triangle Y|, $$ and the correction condition is $$ 2(p+2t+s)