Evaluation of Spectrum Sharing Algorithms for Networks with Heterogeneous Wireless Devices

TL;DR

The paper addresses spectrum sharing in heterogeneous 6G networks with $N$ transmitters and $F$ contiguous bands, where overlapping coverages must use different bands. It evaluates five DSA algorithms built from distinct sorting criteria and uses metrics $FI$, $BU$, $CA$, $BC$, and $TF$ to compare performance via Monte Carlo simulations. The main finding is a trade-off: algorithms prioritizing high-conflict transmitters (e.g., Most-Overlaps, Bandwidth-Coverage) achieve higher $FI$ but often have lower $TF$, while algorithms prioritizing small-resource transmitters ($Least-Bandwidth$, $Least-Coverage$) maximize $TF$ at the cost of $FI$ and $CA$. These results offer guidance for designing DSA in $6$G, illustrating how sorting strategy shapes feasibility, resource utilization, and coverage under limited spectrum.

Abstract

As highlighted in the National Spectrum Strategy, Dynamic Spectrum Access (DSA) is key for enabling 6G networks to meet the increasing demand for spectrum from various, heterogeneous emerging applications. In this paper, we consider heterogeneous wireless networks with multiple 6G base stations (BS) and a limited number of frequency bands available for transmission. Each BS is associated with a geographical location, a coverage area, and a bandwidth requirement. We assume that clients/UEs are within the corresponding BS's coverage area. To avoid interference, we impose that BSs with overlapping coverage areas must use different frequency bands. We address the challenging problem of efficiently allocating contiguous frequency bands to BSs while avoiding interference. Specifically, we define performance metrics that capture the feasibility of the frequency allocation task, the number of BSs that can be allocated within the limited frequency bands, and the amount of resources utilized by the network. Then, we consider five different DSA algorithms that prioritize BSs based on different features -- one of these algorithms is known in the graph theory literature as Welsh-Powell graph colouring algorithm -- and compare their performance using extensive simulations. Our results show that DSA algorithms that attempt to maximize the chances of obtaining a feasible frequency allocation -- which have been widely studied in the literature -- tend to under-perform in all other metrics.

Figures (5)

• Figure 1: Illustration of $N=25$ transmitters located in a two-dimensional region $\mathcal{R}$ of $100\times100$ square meters. Each transmitter is associated with a coverage area. To prevent harmful interference, it is imposed that transmitters with overlapping coverage cannot use the same frequency bands.
• Figure 2: Outcome of a DSA algorithm for the $N=25$ transmitters in Figure \ref{['fig:result_map']} and a total of $F=10$ units of available bandwidth. Each transmitter is allocated its required bandwidth $B_i$ while avoiding interference. Transmitter "3" (identified as "w3" in Figure \ref{['fig:result_map']}) overlaps with "1" and "21." Transmitters "1" and "21," which do not overlap, are assigned bandwidths $f\in\{1,2\}$ and $f\in\{1\}$, respectively, while "3" is assigned $f\in\{3\}$ to avoid interference. Transmitter "19" overlaps with "10," "11," "2," "17," and "12." The only bandwidth available for transmitter "19" is $f\in\{10\}$.
• Figure 3: Performance of networks with increasing number of transmitters $N\in\{5,6,\ldots,30\}$. Unless indicated otherwise, network parameters are set according to Table \ref{['tab:sim_parameters']}. (a), (b), and (c) represent heterogeneous networks with coverage $R_i \sim Uniform(8, 17)$ and bandwidth requirement $B_i \sim Uniform(1, 3)$. (d) represents homogeneous networks with $R_i = 12,\forall i,$ and $B_i = 2,\forall i$. (a) shows the Feasibility Indicator, $FI$, averaged over $500$ simulation runs. (b) shows the Bandwidth Usage, $BU$, averaged over $50$ simulation runs. (c) and (d) show the Total Transmitters while Feasible, $TF$, averaged over $50$ simulation runs. Notice that the best performing DSA algorithms in (a) and (b) are the worse in (c).
• Figure 4: Performance of networks with increasing total available bandwidth $F\in\{5,6,\ldots,15\}$. (a) shows the Total Transmitters while Feasible, $TF$. (b) shows the Coverage Area, $CA$.
• Figure 5: Performance of networks in terms of the Bandwidth-Coverage Product, $BC$, associated with their admissible transmitters. A high $BC$ indicates a DSA algorithm that can allocate transmitters that can cover a large area and can support high quality of service. (a) and (b) represents networks with increasing heterogeneity. (a) has $N=25$ transmitters, each with coverage radius sampled according to $R_i \sim Uniform(8, R_{max})$ with increasing maximum transmitter radius $R_{max}\in\{8,9,\ldots,30\}$. (b) has $N=25$ transmitters, each with required bandwidth sampled according to $B_i \sim Uniform(1, B_{max})$ with increasing maximum transmitter bandwidth $B_{max}\in\{1,2,\ldots,8\}$. Other parameters are set as in Table \ref{['tab:sim_parameters']}.