Throughput Analysis and On-Board Buffer Sizing for Hybrid RF and Optical LEO Satellites Cao-Vien Phung, Thomas Röthig, and Admela Jukan Abstract Low Earth Orbit (LEO) satellite networks are increasingly adopting laser (Free Space Optics, FSO) links to provide high-capacity communications. Although laser inter-satellite links offer high throughput and low latency, RF up- and downlinks remain necessary to maintain connectivity during optical outages caused by adverse atmospheric conditions. In such hybrid link scenarios, satellite buffer design remains a key challenge, since up- and downlink traffic must be buffered and forwarded among satellite nodes. The hybrid RF/FSO scenario requires careful transmission scheduling, especially at envisioned optical transmission rates of 100Gb/s and beyond, making buffer sizing critical under strict onboard energy and weight constraints. Thus, this paper analyzes throughput performance and buffer sizing in hybrid RF/laser satellite networks with finite buffer capacity, interference-aware scheduling, and weather-dependent laser link outage probabilities. Numerical results indicate that laser communications bring significant performance gains. Instead of increasing the transmission power of the satellite to maximize the throughput, we can select a suitable transmission scheduling priority to achieve a maximum throughput, while minimizing the buffer requirement, and lowering packet loss probability under realistic operational conditions and constraints. I Introduction Modern satellite networks increasingly consider laser communication for optical up- and downlinks (LaserCom), as well as laser inter-satellite links (LISLs), driven by ever-increasing capacity requirements [ 1 ] . However, relying solely on laser links remains challenging: weather effects, including clouds, fog, and turbulence, keep optical up- and downlink performance variable. In addition, optical wavelength reuse causes co-channel interference [ 4 ] . Therefore, hybrid FSO/RF links still remain necessary [ 12 ] . Moreover, since current LEO satellite constellations rely on electronic packet processing rather than optical switching, traffic must be routed and buffered onboard before forwarding [ 8 ] . Although such buffering can reduce packet loss during outages or contention [ 3 ] , onboard processing and memory resources remain limited by satellite SWaP (size, weight, power) budgets, making arbitrarily large buffers impractical for high-speed transmission toward hundreds of Gb/s [ 3 ] . In contrast, future optical satellite networks can incorporate transparent all-optical forwarding to reduce electronic processing, lower latency, and improve scalability. Ideally, such systems would operate with little or no buffering or electronic processing. However, fully bufferless operation is not feasible in practice, for a myriad of reasons, most notably because optical feeder links may be intermittently unavailable due to weather effects, while contention can still arise from traffic dynamics and wavelength reuse. As weather-induced outages increase buffering requirements, higher transmit power can reduce the optical link outage probability [ 12 ] , thereby lowering required onboard buffering. However, this shifts the resource burden from storage to power rather, and may still be impractical to SWaP constraints in satellites [ 7 ] . Before a fully bufferless all-optical operation becomes feasible, a practical intermediate step is to minimize onboard buffering while maintaining network throughput performance. This paper investigates the minimum onboard buffer capacity required to maintain throughput performance under adverse weather conditions affecting optical satellite up- and downlinks. To this end, we propose a Markov chain model for the evaluation of throughput and onboard buffer sizing in satellite networks. The model characterizes system performance as a function of finite satellite buffer capacity, transmission scheduling parameter, and link outage probabilities under various weather conditions. The transmission scheduling parameter is optimized to solve outages, reduce required buffer size, decrease packet dropping probability, and improve throughput. Numerical results show that instead of increasing the transmission power of the satellite to maximize the performance, as it is commonly the case, we can select a suitable optical transmission scheduling priority to achieve maximum throughput and minimum buffer size, while lowering the packet dropping probability. The rest of this paper is organized as follows. Section II presents the related work. Section III provides the analysis. Section IV presents performance evaluation. We conclude the paper in Section V . II Related work Paper in [ 12 ] analyzes the probability of an outage link of a weather-dependent hybrid RF/laser link for improved power efficiency. We extend this paper with an analysis of interference-avoiding transmission scheduling strategies to solve the outage problem, while maximizing the throughput and minimizing the buffer. Papers [ 2 , 10 ] analyze the performance with interference-avoiding transmission scheduling strategies in a general wireless network with a limit for the relay node buffer, using the Markov chain model. Paper [ 6 ] builds an analytical model for mixed laser/RF UAV-aided mobile relaying with a buffer, which we use in this paper to analyze foggy weather. However, neither of these papers analyzes outages and throughput performance in the same framework, which is our goal. The models in [ 5 , 3 ] are applied for satellite networks with limited buffer/cache used to retransmit error/lost bursts/chunks. These studies focus on selecting a data burst/chunk size transferred during its burst/chunk time, which is different from our study. It should be noted that many of the related work listed are specifically referred in the analysis, whenever re-used from other models. In a nutshell, the novelty of our paper is that it combines the aspects of traffic, satellite buffer limit, transmission scheduling parameters, and outage link parameters under different weather effects (thin cloud, fog, and rain) which as combination is novel, and required in hybrid RF/FSO scenario. Figure 1: Reference scenario under different weather effects. Table I: List of main parameters. Notation Meaning A_{\gamma} Average queue length [bits], A_{\gamma}=\{A_{C}:cloud,A_{F}:fog,A_{R}:rain\} . C_{\gamma} Capacity [bps], C_{\gamma}=\{C_{SG}^{f,C}:cloud\;with\;laser,C_{SG}^{f,R}: rain,C_{SG}^{r,C}:cloud\;with\;RF,C_{SG}^{r,F}:fog,C_{SS}:LISL\} . L , L_{b} Buffer length limit of satellite SA I , L [packets], L_{b} [bits]. O_{t} Total time observed in the weather using RF link [seconds]. P_{\gamma} Average packet dropping probability, P_{\gamma}=\{P_{C}:cloud,P_{F}:fog,P_{R}:rain\} . P(i) State probability of i packets in the buffer. P_{l} Transition probability from state i\in(0,L) to itself. P_{l}^{{}^{\prime}} Transition probability from state i=L to itself. P_{SA^{I}} Transition probability from state i\in(0,L] to state i-1 . P_{SA^{I}}^{{}^{\prime}} Probability that SA I sends unsuccessfully data to GS. P_{SA^{s}} Transition probability from state i\in(0,L) to state i+1 . P_{SA^{s}}^{{}^{\prime}} Probability that SA s sends unsuccessfully data to SA I . T_{SG}^{o} Timeout for transmission via SA I -GS link [seconds]. T_{SS}^{o} Timeout for transmission via SA s -SA I link [seconds]. X Average size of a packet [bits]. W_{\gamma} Probability W_{\gamma}=\{W_{C}:cloud,W_{R}:rain,W_{F}:fog\} . \alpha SA I priority of accessing medium (real value). \rho_{\gamma} Outage probability, \rho_{\gamma}=\{\rho_{SG,C}:cloud, \rho_{SG,F}:fog,\rho_{SG,R}:rain,\rho_{SS}:LISL\} . \tau_{\gamma} Average throughput [bps], \tau_{\gamma}=\{\tau_{C}:cloud,\tau_{F}:fog,\tau_{R}:rain\} . III Buffer and throughput analysis III-A Reference scenario Our reference scenario in Fig. 1 includes one LEO Satellite source (SA s ), one intermediate LEO Satellite (SA I ), and one GS (destination node), where SA I always connects with SA s , either directly or by handover, under the assumption of a satellite grid, and SA I can serve one or more GSs during its movements. For simplicity and without loss of generality, assume that SA I can connect with one SA s and one GS only. Table I summarizes the main notations in the paper. Table II: Link selections between SA I and GS [ 12 ] . Weather condition SA I to GS link Thin cloud Both laser and RF links (dual channels) Rain laser link Fog RF link The system always has packets to send from SA s . The capacity of the SA s -SA I LISL is given by: C_{SS}=W_{f}\cdot log_{2}(1+SNR_{SS}), (1) where W_{f} is the bandwidth and SNR_{SS} is the signal-to-noise ratio [ 9 ] . The SA s -SA I laser link has the probability of outage \rho_{SS} causing a timeout time T^{o}_{SS} [ 3 ] . After a timeout, any unsuccessful packet can be resent, akin to [ 3 ] . SA I does not generate packets itself, as it only relays packets received from SA s . The SA I -GS weather-link selections are summarized in Table II . For thin cloud, the identical rate information is transmitted by both RF and laser channels in parallel [ 12 ] , where the selection combining technique is to combine received signals at GS to maximize performance. For simplicity and without loss of generality, we assume that there exists only one link (either laser or RF) whose performance is calculated on average between laser and RF. Let \rho_{SG,C} , \rho_{SG,R} , and \rho_{SG,F} be the probability of outage of SA I -GS links in thin cloud, rain, and fog, respectively, [ 12 ] , causing the timeout time T^{o}_{SG} . Note that due to optical interference in the same frequency band [ 4 ] , transmitting data to GS and receiving data from SA s at SA I for weather conditions using SA I -GS laser links cannot be simultaneously performed, whereas SA I can simultaneously transmit data to GS for weather conditions using RF and receive data from SA s using laser. The SA I -GS link capacities C_{SG}^{f,C} , C_{SG}^{f,R} , C_{SG}^{r,C} , C_{SG}^{r,F} are for thin cloud with laser link, rain, thin cloud with RF link, and fog, respectively, which are similar to Eq. ( 1 ), where their SNRs are referred to [ 12 ] . Assume that the switching time between laser and RF is negligible, including signal processing at SA I . For weather using SA I -GS laser links, SA s is scheduled to send one packet of X bits on average to SA I during its access. Let L be the SA I buffer limit in packets ( L_{b} in bits) [ 3 ] . Each packet in the buffer occupies one unit. We define \alpha "priority" (real value) for accessing the transmission medium from SA I to GS, i.e., \alpha times of transmission opportunities, whereas SA s competes to get one time for accessing the transmission medium from SA s to SA I . SA I and SA s will not compete for access, if the SA I buffer is empty. If the number of packets stored in the queue is equal to L , SA I will drop any new packet that arrives at its buffer. When SA I wins a transmission chance, it can transport a packet to GS, i.e., remove a packet from its buffer. In case of weather conditions using SA I -GS RF, SA I does not use the priority \alpha because SA s -SA I links are optical, i.e., they do not interfere. III-B Thin cloud weather model III-B 1 Queue length analysis (thin cloud) Lemma 1 . The average queue length of the SA I buffer in the thin cloud using SA I -GS laser/RF link can be given as follows: A_{C}=\frac{1}{2}(A_{C,f}+A_{C,r}), (2) where A_{C,f} and A_{C,r} are the average queue length of SA I if using SA I -GS laser and SA I -GS RF, respectively. A_{C,f} and A_{C,r} are calculated by using Lemmas 4 and 5 , respectively. Figure 2: Embedded Markov chain model applied. With cloud (SA I -GS laser), we design a Markov chain (Fig. 2 ) to obtain the throughput, queue length of buffered packets, and packet dropping probability at SA I , where each state i\in[0,L] is the number of buffered packets in the SA I queue (queue state). Observations in the Markov chain occur before each packet transmission from SA s or SA I . Let P_{0\rightarrow 1} and P_{0\rightarrow 0} be the transition probability from state i=0 (empty buffer) to state i=1 and to itself, respectively. Since the SA I buffer is empty, the transmission chance is only from SA s which increases one unit in the queue. So, we get P_{0\rightarrow 1}=1-\rho_{SS} in case of which SA s sends successfully one packet to SA I , whereas we get P_{0\rightarrow 0}=\rho_{SS} in case of which SA s sends unsuccessfully one due to SA s -SA I link outage. Let P_{SA^{s}} be the transition probability that nonempty queue state i\in(0,L) to i+1 , i.e., queue increases one, in case of transmission chance from SA s occupied. As the SA I buffer is not empty, its transmission competition chance is " \alpha priority" over SA s . Hence, with the transmission chance of SA s occupied and SA s -SA I link outage probability \rho_{SS} , probability P_{SA^{s}} that the nonempty SA I queue increases one packet (SA s sends successfully one to SA I ) is given by: P_{SA^{s}}=\frac{1-\rho_{SS}}{1+\alpha}, (3) whereas the probability P_{SA^{s}}^{{}^{\prime}} that the nonempty SA I queue keeps unchanged (SA s sends unsuccessfully one to SA I ) is: P_{SA^{s}}^{{}^{\prime}}=\frac{\rho_{SS}}{1+\alpha}. (4) Let P_{SA^{I}} be the transition probability that nonempty queue state i\in(0,L] to i-1 , i.e., queue decreases one packet, in case of transmission chance from SA I occupied with the " \alpha priority". With \rho_{SG,C} of SA I -GS laser link outage probability, the probability P_{SA^{I}} that the nonempty SA I queue decreases one packet (SA I sends successfully one to GS) is given by: P_{SA^{I}}=\frac{\alpha(1-\rho_{SG,C})}{1+\alpha}, (5) whereas the probability P_{SA^{I}}^{{}^{\prime}} that the nonempty SA I queue keeps unchanged (SA I sends unsuccessfully one to GS) is: P_{SA^{I}}^{{}^{\prime}}=\frac{\alpha\cdot\rho_{SG,C}}{1+\alpha}. (6) With Eqs. ( 4 ), ( 6 ), the transition probability P_{l} from state i\in(0,L) to itself, i.e., the probability that the nonempty SA I queue keeps unchanged (SA s or SA I sends unsuccessfully one packet to SA I or GS, respectively) is given as follows: P_{l}=P_{SA^{s}}^{{}^{\prime}}+P_{SA^{I}}^{{}^{\prime}}=\frac{\rho_{SS}+\alpha\cdot\rho_{SG,C}}{1+\alpha}. (7) At i=L , any new packet that arrives at SA I when SA s sends data with the probability P_{SA^{s}} will be dropped. With Eqs. ( 4 ), ( 6 ), ( 7 ), the transition probability from i=L to itself is given as follows: P_{l}^{{}^{\prime}}=P_{SA^{s}}+P_{l}=\frac{1+\alpha\cdot\rho_{SG,C}}{1+\alpha}. (8) Lemma 2 . The Markov chain is finite, irreducible, aperiodic. The proof is provided in Appendix -1 . Lemma 3 . Let P(i) be the state probability that the SA I buffer has i packets. P(i) in thin cloud (SA I -GS laser link) is: P(0)=\frac{1}{1+(1-\rho_{SS})\sum_{i=1}^{L}\frac{(P_{SA^{s}})^{i-1}}{(P_{SA^{I}})^{i}}}, (9) P(i)=\frac{(P_{SA^{s}})^{i-1}}{(P_{SA^{I}})^{i}}(1-\rho_{SS})P(0),\forall i\in[1,L]. (10) The proof is provided in Appendix -2 . Lemma 4 . The average queue length of SA I buffer in the thin cloud using SA I -GS laser link can be given as follows: A_{C,f}=X(1-\rho_{SS})P(0)\sum_{i=1}^{L}i\frac{(P_{SA^{s}})^{i-1}}{(P_{SA^{I}})^{i}}, (11) where P(0) is referred to Lemma 3 . Lemma 4 proves by using X\sum_{i=1}^{L}i\cdot P(i) and Lemma 3 . Lemma 5 . The queue length of SA I buffer (i.e., using SA I -GS RF) is the difference between the capacities of SA s -SA I laser and RF links with observed time O_{t} s. So, it is given by: \begin{cases}\begin{split}&A_{C,r}=\frac{1}{O_{t}+1}\sum_{i=0}^{O_{t}}min\left(L_{b},i\cdot[\beta-\phi]\right),\text{if }\beta>\phi,\\ &A_{C,r}=0,\;\text{if }\beta\leq\phi,\end{split}\end{cases} (12) where \beta=(1-\rho_{SS})\cdot C_{SS} is the amount of arriving data at SA I from SA s and \phi=(1-\rho_{SG,C})\cdot C^{r,C}_{SG} is the amount of data released from SA I to GS. III-B 2 Packet dropping probability analysis (thin cloud) We define the packet dropping probability to be packets that arrives at SA I where refuses buffering them due to full buffer. Lemma 6 . The packet dropping probability at the SA I buffer in the thin cloud using SA I -GS laser/RF link is given by: P_{C}=\frac{1}{2}(P_{C,f}+P_{C,r}), (13) where P_{C,f} and P_{C,r} are the packet dropping probability at SA I buffer in the thin cloud if using SA I -GS laser and RF links, respectively. P_{C,f} and P_{C,r} are from Lemmas 7 , 8 . Lemma 7 . The packet dropping probability at SA I in the thin cloud using SA I -GS laser link is given as follows: P_{C,f}=\left(\frac{P_{SA^{s}}}{P_{SA^{I}}}\right)^{L}(1-\rho_{SS})P(0), (14) where P(0) is referred to Lemma 3 . Lemma 7 is proved by using P_{SA}\cdot P(L) and Lemma 3 . Lemma 8 . The packet dropping probability at SA I in the thin cloud using SA I -GS RF is given as follows: \begin{split}P_{C,r}=1-\left[\frac{1}{O_{t}}\sum_{i=1}^{O_{t}}min\left(1,\frac{S^{z}_{e}+\phi}{\beta}\right)\right],\end{split} (15) where the remaining SA I buffer space in bits at the observed time z=i-1 (previously observed time i ) is given as follows: \begin{cases}\begin{split}&S_{e}^{z}=L_{b}-min\left(L_{b},z\cdot[\beta-\phi]\right),\text{if }\beta>\phi,\\ &S_{e}^{z}=L_{b},\;\text{if }\beta\leq\phi,\end{split}\end{cases} (16) All parameters in Lemma 8 can be referred to Lemma 5 . III-B 3 Throughput analysis We define the average throughput to be the ratio of the average amount of data per transmission slot that SA I forwards to the average transmission slot length. Lemma 9 . The average throughput in the thin cloud using SA I -GS laser/RF link can be given as follows: \tau_{C}=\frac{1}{2}(\tau_{C,f}+\tau_{C,r}), (17) where \tau_{C,f} and \tau_{C,r} are the throughput in the thin cloud if using SA I -GS laser and SA I -GS RF link, respectively. \tau_{C,f} and \tau_{C,r} are from Lemma 10 and Lemma 11 . Lemma 10 . The average network throughput in the thin cloud using SA I -GS laser link is given as follows: \tau_{C,f}=\frac{D_{C,f}}{T_{C,f}}, (18) where the average amount of data per transmission slot that SA I forwards can be given as follows: D_{C,f}=[1-P(0)]P_{SA^{I}}\cdot X, (19) and the average transmission slot length is given as follows: \begin{split}T_{C,f}&=P(0)\cdot(1-\rho_{SS})\frac{X}{C_{SS}}+P(0)\cdot\rho_{SS}\cdot T^{o}_{SS}\\ &+[1-P(0)]P_{SA^{s}}\frac{X}{C_{SS}}+[1-P(0)]P_{SA^{s}}^{{}^{\prime}}\cdot T^{o}_{SS}\\ &+[1-P(0)]P_{SA^{I}}\frac{X}{C^{f,C}_{SG}}+[1-P(0)]P_{SA^{I}}^{{}^{\prime}}\cdot T^{o}_{SG},\end{split} (20) where P(0) is referred to Lemma 3 . The proof is provided in Appendix -3 . Lemma 11 . The average network throughput in the thin cloud using SA I -GS RF link is given as follows: \tau_{C,r}=min([1-P_{C,r}]\beta,\phi), (21) where all parameters are referred to Lemma 5 , and P_{C,r} is from Lemma 8 . III-C Rain weather model With the rain (SA I -GS laser link), the average queue length A_{R} , packet dropping probability P_{R} , and throughput \tau_{R} can be similarly calculated using Lemmas 4 , 7 , and 10 . III-D Foggy weather model In foggy weather (SA I -GS RF link), the average queue length A_{F} , packet dropping probability P_{F} , and throughput \tau_{F} can be similarly calculated using Lemmas 5 , 8 , and 11 . III-E Combined weather models As the satellite moves, it can experience different weather effects. Let W_{C} , W_{R} , and W_{F} be the probability of thin cloud, rain, and fog, respectively. Theorem 1 . The queue length of SA I under three weather effects (thin cloud, rain, and fog) combined is given as follows: A_{t}=W_{C}\cdot A_{C}+W_{R}\cdot A_{R}+W_{F}\cdot A_{F}. (22) Theorem 2 . The packet dropping probability at SA I under the thin cloud, rain, and fog combined is given as follows: P_{t}=W_{C}\cdot P_{C}+W_{R}\cdot P_{R}+W_{F}\cdot P_{F}. (23) Theorem 3 . The network throughput under three weather effects (thin cloud, rain, and fog) combined is given as follows: \tau_{t}=W_{C}\cdot\tau_{C}+W_{R}\cdot\tau_{R}+W_{F}\cdot\tau_{F}. (24) (a) (b) (c) Figure 3: Satellite system performance over different SA I priorities \alpha , where the SA I buffer limit is fixed at L=100 packets. (a) (b) (c) Figure 4: Satellite system performance over different SA I buffer limits L , where the SA I priority is fixed at \alpha=1 . IV Performance evaluation We start the evaluation with the system configuration assumptions. The satellite and GS altitudes are set at 500 km and 0.8 km, respectively. The average SA s -SA I distance is 500 km. The bandwidth W_{f} for the optical and W_{r} for RF is 500 MHz and 50 MHz, respectively. The RF frequency is 40 GHz, while the laser wavelength is 1550 nm. The satellite and GS gains are 52 dBi. The evaluation angle is 30^{o} . The transmission power P^{t}_{SA^{s}} of SA^{s} is 0.7 watt. The timeouts T^{o}_{SS} for SA s -SA I link and T^{o}_{SG} for SA I -GS link are 0.01 s. Each packet X has 20 Mbits on average. Assume that the average outage probability of SA s -SA I link is small, e.g., 0.0001 . For fairness, assume that the probabilities of the different weather types ( W_{C} , W_{R} , and W_{F} ) are equal. The observed time O_{t} in case of weather, such as fog, only needs a few dozen seconds to get convergent results which were also proved in [ 10 ] , while the weather effects may be occurred in a longer duration. The remaining parameters are selected according to [ 12 , 11 , 9 ] . We focus on evaluating the throughput (Theorem 3 ), queue length of SA I buffer (Theorem 1 ), and packet dropping probability (Theorem 2 ) under three weather effects (thin cloud, rain, and fog) combined. In Fig. 3 , we consider a SA I buffer with L=100 packets, different SA I transmission scheduling priorities \alpha=[0.5,10] in step of 0.5 ( \alpha is only used for SA I -GS laser), and transmission power of SA I P^{t}_{SA^{I}}=\{-20,-10,0\} dBm. As we analyze results under combined weather effects. The smaller the P^{t}_{SA^{I}} value is, the worse the performance is because the probability of SA I -GS link outage increases with decreasing P^{t}_{SA^{I}} . Also, the throughput in Fig. 3a increases with increasing \alpha . This is because the buffered packets can be removed faster with a higher \alpha (average queue length in Fig. 3b and packet dropping probability in Fig. 3c decrease with increasing \alpha ). At a fixed transmission power P^{t}_{SA^{I}} , we can increase the performance by adjusting the priority \alpha without needing an increment of P^{t}_{SA^{I}} . However, the throughput achieves a maximum at \alpha\geq 1.5 and configuring \alpha larger than 1.5 is unnecessary because the packet dropping probability at \alpha\geq 1.5 is constant. The reason is that setting a SA I priority \alpha higher than 1.5 will occupy more redundant transmission chance over SA s , leading to sometime less data transmitted from SA s due to using an inefficient frequency resource. In Fig. 3 , in case of the weather using SA I -GS RF link, the performance does not depend on the SA I priority \alpha . The performance is constant after observed time O_{t} because the arrival data rate of SA s -SA I laser link is much larger than the released data rate of SA I -GS RF link. Thus, the queue length of SA I buffer is always approximately full, and the packet dropping probability is only the ratio of the amount of released data to the amount of arrival data. Hence, the variant performance over different \alpha values mainly comes from the case of the weather using SA I -GS laser link. Due to after \alpha\geq 1.5 , the performance is constant. Thus, to get a much better quality of service, we need to sacrifice the consumed energy of satellite by increasing P^{t}_{SA^{I}} . In addition, the average queue length and packet dropping probability are relatively high, even increasing P^{t}_{SA^{I}} and \alpha . This is due to bad performance by using SA I -GS RF link which is independent with using the priority \alpha . Thus, the satellite systems prioritize using the laser technology, while the RF technology is used as the backup [ 10 ] . We now show the benefit of using the laser links. Assume that we only consider the foggy weather using SA I -GS laser link with adjusting \alpha . The system achieves a maximum performance at \alpha=1.5 : The throughput achieves 3.8768 Gbps at P^{t}_{SA^{I}}=-20 dBm, 4.9615 Gbps at P^{t}_{SA^{I}}=-10 dBm, and 6.1077 Gbps at P^{t}_{SA^{I}}=0 dBm; the queue length is optimally minimized at 0.0936 Gbits for P^{t}_{SA^{I}}=-20 dBm, at 0.0740 Gbits for P^{t}_{SA^{I}}=-10 dBm, and at 0.0647 Gbits for P^{t}_{SA^{I}}=0 dBm; the packet dropping probability achieves zero for all cases. Next, we consider a SA I buffer with L=[10,100] packets in step of 10 , SA I priority \alpha=1 , and P^{t}_{SA^{I}}=\{-20,-10,0\} dBm. The throughput shown (Fig. 4a ) increases with increasing L because the SA I buffer with a higher L value can save more data which is proved via the average queue length (Fig. 4b ), so this leads to lower packet dropping probabilities (Fig. 4c ). However, the total satellite system performance slowly increases with increasing L because of the inefficiency in case of the weather using SA I -GS RF link as analyzed above. V Conclusion We proposed an analytical framework based on Markov chain model to calculate the throughput and on-board buffering requirements for hybrid RF/FSO networks. The analytical results also show that instead of increasing transmission power, which increases energy and weight requirements, a suitable optical transmission scheduling priority can be used to maximize throughput and minimize electronic buffer size. We observed best results when using laser downlinks, resulting in a small buffer and zero packet dropping probability. RF improves the FSO reliability, but it requires comparably larger buffers, also due to reduced capacity. Our analysis also showed that even with smaller power, we can minimize the buffer, which carries potential for future all-optical transparent network solutions. - 1 Proof of Lemma 2 As the state space in the Markov chain has L states, it is finite. In addition, this Markov chain is irreducible because we observe Fig. 2 that any state i\in[0,L] can be reachable from any other state j\in[0,L]\setminus i with a non-zero probability. Finally, we prove that if any state from this Markov chain is aperiodic, then it is aperiodic. Any state i with the period d(i) has the greatest common denominator from all integer values larger than 0 , whereby there exists a transition probability larger than 0 from any state i traversing to the other states, and then going back to itself. In this Markov chain, there always exists a state i with a non-zero self-transition probability, d(i)=1 . Thus, the state i is aperiodic. - 2 Proof of Lemma 3 In Fig. 2 , the balance equations can be given as follows: P(1)=\frac{1-\rho_{SS}}{P_{SA^{I}}}P(0);P(i+1)=\frac{P_{SA^{s}}}{P_{SA^{I}}}P(i),\forall i\in[1,L). (25) With Eqs. ( 25 ), and \sum_{i=0}^{L}P(i)=1 , we get Lemma 3 . - 3 Proof of Lemma 10 Based on Fig. 2 , if the buffer is empty with the state probability P(0) , the transmission probability of SA s without competing with SA I that sends successfully one packet to SA I is 1-\rho_{SS} with the transmission slot length \frac{X}{C_{SS}} , i.e., P(0)\cdot(1-\rho_{SS})\frac{X}{C_{SS}} . In case of empty buffer, the timeout time T^{o}_{SS} for the SA s -SA I laser link due to its outage probability \rho_{SS} causing packets unsuccessfully sent is P(0)\cdot\rho_{SS}\cdot T^{o}_{SS} . If the buffer is nonempty with the probability P(\bar{0})=1-P(0) , the transmission probability of SA s with competing with SA I that sends successfully one to SA I is P_{SA^{s}} with the transmission slot length \frac{X}{C_{SS}} , i.e., P(\bar{0})\cdot P_{SA^{s}}\frac{X}{C_{SS}} . Also, in case of nonempty buffer, the transmission probability of SA s with competing with SA I that sends unsuccessfully one to SA I is P_{SA^{s}}^{{}^{\prime}} with timeout time T^{o}_{SS} , i.e., P(\bar{0})\cdot P_{SA^{s}}^{{}^{\prime}}\cdot T^{o}_{SS} . Similarly as explained above, we have the transmission slot length if SA I wins transmission chances, i.e., P(\bar{0})\cdot P_{SA^{I}}\frac{X}{C^{f,C}_{SG}}+P(\bar{0})\cdot P_{SA^{I}}^{{}^{\prime}}\cdot T^{o}_{SG} . So, we get Eq. ( 20 ). If the buffer is nonempty, the transmission probability of SA I with competing with SA s that sends successfully one packet ( X bits) to GS is P_{SA^{I}} , i.e., Eq. ( 19 ).