3.1. Network Topology Model
We model an LEO satellite network as a directed graph
, where
V is the set of satellites and
E is the set of all possible links between satellites. In order to cope with the dynamic changes of topology, this paper adopts the virtual topology approach. The virtual topology method divides the motion period
T of the satellite network into
time slots. Each time slot
i corresponds to a topology snapshot [
17,
18], indicating that there are no changes in the topology snapshot during time slot
i. For simplicity, we assume that all time slots have the same length. However, the algorithm designed in this paper can also be used directly in the cases of non-equal time slots.
Since satellites move continuously over time, a link may be disconnected during time slot i. Therefore, we use to indicate whether link e exists in the i-th time slot. Additionally, the propagation delay of links varies over time, so we use to represent the maximum propagation delay of link e in time slot i.
In this paper, we mainly focus on scheduling time-sensitive (TS) flows. The set of TS flows to be scheduled is denoted as F, where the j-th TS flow is represented by . Let and represent the bandwidth requirement and maximum transmission delay constraint of flow .
3.2. Problem Formulation
In this work, our primary goal is to meet the propagation delay constraints and minimize queuing delay for TS services in LEO satellite networks through optimized routing. Furthermore, due to the highly dynamic topology of LEO satellite networks, we need to consider minimizing the control overhead caused by frequent routing changes. Therefore, in modeling the routing problem, we have incorporated service propagation delay requirements and network bandwidth constraints, while load balancing and routing handover overhead are set as optimization objectives. Moreover, our proposed routing algorithm can also easily incorporate other optimization goals, such as minimizing energy consumption, through simple extensions. Since the routing optimization problem for TS flows in LEO satellite networks needs to consider multiple topology snapshots simultaneously, it is much more complex than the routing optimization problem in static topologies. There are usually two modeling approaches for network routing optimization problems: node-link and link-path [
26]. In the link-path model, the candidate paths for each flow are given in advance, and thus we can control the complexity of the model by adjusting the number of candidate paths. However, the node-link model requires the direct computing of all paths for the TS flows, thus introducing more variables and constraints, which makes the node-link model difficult to solve in large-scale networks. To reduce the complexity of the model, we choose the link-path approach to model the problem in this paper.
For the sake of clarity, we provide a comprehensive list of the notations used in
Table 1. In the following text, we will first introduce the routing constraints and then give the optimization objective.
Using
Figure 3, we can illustrate the decision variables more clearly. There are two consecutive time slots, slot
i and slot
; each node represents a satellite, and solid lines denote available links between two satellites. We observe that there is a change in the state of
, while
remain in a connected state, and
remains unconnected. Given that
indicates whether link
e exists in time slot
i, it can be easily observed that
. Then, we assume there is a TS flow
from source node
S to destination node
D, and in time slot
i, there are two paths, Path 1 and Path 2, from the candidate paths set
. We can observe that Path 2 includes link
and excludes link
, so
. Then, we assume that in slot
i and
, we have both chosen Path 1, that is,
and
. However, a path handover occurs due to the fact that the links of the paths selected in the two slots are not entirely coincident, so
will be set to 1.
Firstly, the maximum delay of the path chosen for each TS flow in each topology snapshot should meet the flow’s delay constraint.
If flow
uses path
k in time slot
i, then every link traversed by path
k should exist in time slot
i, i.e., we have the following constraint:
To avoid packet reordering, we assume that each TS flow can only use one path within each time slot. Therefore, the decision variable
needs to satisfy the following constraint:
Furthermore, to avoid introducing significant unpredictable delays due to traffic overload, we require that the total size of TS flows carried on each link should be smaller than its capacity.
After introducing the constraints that TS flows need to satisfy, we now turn our attention to the optimization objective of the problem. As introduced in
Section 2, we mainly consider two optimization objectives in this problem. First of all, we believe that the future LEO satellite networks will use SDN technology to control flow routing in a fine-grained manner. Therefore, in order to reduce the control overhead of SDN, we need to minimize the total number of path handovers. Secondly, network congestion can lead to a significant increase in packet queuing delay. Therefore, to better guarantee the transmission delay requirements of TS flows, we need to achieve network load balancing.
Firstly, we address the optimization of the number of path handovers. For any given flow, within two consecutive time slots
and
i, a path handover occurs if the routing path changes between these two time slots. Specifically, a path handover is counted if any link used by the flow’s path in time slot
is different from the links used in time slot
i. Furthermore, even if multiple links change between the two time slots, only one path handover is counted. To model the path handovers, we introduce a binary variable
which indicates whether flow
j changes its path during time slot
to
i. The following constraint are added to relate
with the path selection variables
:
To satisfy the constraints of the ILP formulation, we expand the above constraint into the following two constraints:
To quantitatively analyze the number of path switches, we let
denote the average number of path switches per time slot for each flow, which is defined as:
Then, to indicate the degree of the load balancing of the satellite network, we define
as the maximum link utilization in the network. From the right-hand side of Equation (
9), we can calculate the link utilization for each link within all time slots, where
represents the maximum value. Therefore, by utilizing Equation (
9) we ensure that
is greater than the link utilization of each individual link within all time slots.
Finally, taking into account both the number of path handovers and network load balancing, the ILP formulation of our problem is:
So far, we have successfully modeled the routing optimization problem for TS flows in LEO satellite networks as an integer linear programming (ILP) model. While solving this model can provide the optimal routing solution for TS flows, the complexity of solving ILP models is exponential, which is not feasible for large-scale LEO satellite networks. Furthermore, it can be easily proven that this problem is an NP-complete problem since even when simplifying the problem to a single time slot scenario, it remains NP-hard [
27]. To effectively solve this problem, we propose an approximate algorithm called the longest continuous path (LCP), which strikes a good balance between solution speed and performance. The details of the LCP algorithm will be discussed in the next section.