Abstract:Understanding the maximum size of a code with a given minimum distance is a major question in computer science and discrete mathematics. The most fruitful approach for finding asymptotic bounds on such codes is by using Delsarte's theory of association schemes. With this approach, Delsarte constructs a linear program such that its maximum value is an upper bound on the maximum size of a code with a given minimum distance. Bounding this value can be done by finding solutions to the corresponding dual linear program. Delsarte's theory is very general and goes way beyond binary codes. In this work, we provide universal bounds in the framework of association schemes that generalize the Elias-Bassalygo bound, which can be applied to any association scheme constructed from a distance function. These bounds are obtained by constructing new solutions to Delsarte's dual linear program. We instantiate these results and we recover known bounds for $q$-ary codes and for constant-weight binary codes. Our other contribution is to recover, for essentially any $Q$-polynomial scheme, MRRW-type solutions to Delsarte's dual linear program which are inspired by the Laplacian approach of Friedman and Tillich instead of using the Christoffel-Darboux formulas. We show in particular how the second linear programming bound can be interpreted in this framework.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is, in coding theory, how to find the size of the largest codeword set with a given minimum distance. Specifically, the goal of the paper is to provide general bounds by constructing new solutions to the Delsarte dual linear program, and these bounds can be applied to any association scheme constructed from a distance function. This includes:
1. **Providing general bounds**: The paper proposes general bounds within the framework of the association scheme, which generalize the Elias - Bassalygo bounds and are applicable to any association scheme constructed from a distance function.
2. **Recovering known bounds**: By instantiating these results, the paper recovers the known bounds for q - ary codes and constant - weight binary codes.
3. **New dual linear programming solutions**: For almost all Q - polynomial schemes, the paper recovers the MRRW - type dual linear programming solutions inspired by Friedman and Tillich's Laplacian method, instead of using the Christoffel - Darboux formula.
### Specific problems and methods
- **Size of the largest codeword set**: Understanding the size of the largest codeword set with a given minimum distance is an important problem in computer science and discrete mathematics. Delsarte constructed a linear program through the theory of association schemes, and its maximum value is an upper bound on the size of the largest codeword set with a given minimum distance.
- **Dual linear programming**: By finding the solution to the corresponding dual linear program, this value can be bounded. Delsarte's theory is very general and is not limited to binary codes.
- **Construction of new solutions**: The paper provides more general bounds than existing methods by constructing new dual linear programming solutions. These new solutions are applicable to any association scheme constructed from a distance function, thus generalizing the Elias - Bassalygo bounds.
### Main contributions
1. **General bounds**: Provided general bounds applicable to any association scheme, which generalize the Elias - Bassalygo bounds.
2. **Recovering known bounds**: By instantiating these general bounds, recovered the known bounds for q - ary codes and constant - weight binary codes.
3. **New MRRW - type solutions**: For almost all Q - polynomial schemes, the paper found MRRW - type dual linear programming solutions inspired by Friedman and Tillich's Laplacian method.
### Mathematical formulas
- **Elias - Bassalygo bounds**:
\[
R(\delta) \leq 1 - h(\alpha)+R(\delta, \alpha)
\]
where \(h(x)=-x \log_2 x-(1 - x) \log_2(1 - x)\) is the binary entropy function.
- **MRRW first bound**:
\[
R(\delta) \leq R_{\text{MRRW1}}(\delta)=h\left(\frac{1}{2}-\frac{1}{2} \sqrt{\delta(1 - \delta)}\right)
\]
- **MRRW second bound**:
\[
R(\delta) \leq R_{\text{MRRW2}}(\delta)=\max _{0 \leq \alpha \leq \frac{1}{2}}\left\{1 - h(\alpha)+R_{\text{MRRW}}(\alpha, \delta)\right\}
\]
where \(R_{\text{MRRW}}(\alpha, \delta)=\frac{1}{2}\left(1-\sqrt{1 - 4\left(\sqrt{\alpha(1 - \alpha)}-\delta(1 - \delta)-\delta\right)^2}\right)\).
### Conclusion
The paper provides general bounds applicable to a wide range of association schemes by constructing new dual linear programming solutions, which not only generalize...