带回退的缓冲Aloha算法的稳定性和吞吐量

Stability Region of a Slotted Aloha Network with K-Exponential Backoff

Stability region of random access wireless networks is known for only simple network scenarios. The main problem in this respect is due to interaction among queues. When transmission probabilities during successive transmissions change, e.g., when exponential backoff mechanism is exploited, the interactions in the network are stimulated. In this paper, we derive the stability region of a buffered slotted Aloha network with K-exponential backoff mechanism, approximately, when a finite number of nodes exist. To this end, we propose a new approach in modeling the interaction among wireless nodes.…

Buffered Aloha with K-Exponential Backoff -- Part II: Delay Analysis

This paper presents the delay analysis for buffered Aloha networks with K-Exponential Backoff. Mean access delay and mean queueing delay are derived and demonstrated via the examples of Geometric Retransmission (K=1) and Exponential Backoff (K=infinity). The comparison shows that higher delay is incurred with Geometric Retransmission when the aggregate input rate is small, and the delay gap is enlarged as the number of nodes n increases. With a high traffic input rate, however, the delay performance with Exponential Backoff severely deteriorates. The mean queueing delay will be unbounded if th…

Buffered Aloha with K-Exponential Backoff -- Part I: Stability and Throughput Analysis

This two-part paper series studies the performance of buffered Aloha networks with K-Exponential Backoff collision resolution algorithms. Part I focuses on stability and throughput analysis and Part II presents the delay analysis. In Part I, the buffered Aloha network is modeled as a multi-queue single-server system. We adopt a widely used approach in packet switching systems to decompose the multi-queue system into independent first-in-first-out (FIFO) queues, which are hinged together by the probability of success of head-of-line (HOL) packets. A unified method is devised to tackle the stabi…

Aloha网络的有界平均延迟吞吐量和非饥饿条件

Bounded Mean-Delay Throughput and Non-Starvation Conditions in Aloha Network

本文考虑了在时隙Aloha网络中，指数退避协议下保证有界平均排队延迟和非饥饿的要求。众所周知，具有大量节点的时隙Aloha系统的最大可能吞吐量是1/e = 0.3679。实际上，1/e的饱和吞吐量可以用指数回退因子r = e/（e-1）=1.5820来实现。在大多数先前的工作中，以及在许多实际的多址接入网络（诸如以太网和WiFi）中，假定二进制退避因子r = 2。对于时隙Aloha，r = 2时的饱和吞吐量0.3466合理地接近1/e的最大值，并且很难反对在系统中采用r = 2。然而，本文表明，如果平均排队延迟是有界的，那么当r = 2时，可持续的吞吐量仅为0.2158，从1/e急剧下降41%。幸运的是，在有界平均延迟要求下的最佳设置r = 1.3757允许我们实现0.3545的可持续吞吐量，相对于1/e的损失仅小于4%。一般的结论是，r的值可能会显着影响排队延迟性能。除了分析平均排队延迟，本文还深入研究了饥饿现象，其中一些节点被剥夺服务的一段时间，而其他节点占用系统。具体来说，我们提出了一个定量的定义饥饿，并表明，保证有界平均延迟和非饥饿操作的条件是相同的，从而统一这两个概念。最后，我们表明，当平均延迟大，饥饿发生时，从模拟实验中得到的性能结果可能不会收敛。本文对此问题进行了定量讨论。

On the Distributed Construction of a Collision-Free Schedule in Multi-Hop Packet Radio Networks

This paper introduces a protocol that distributively constructs a collision-free schedule for multi-hop packet radio networks in the presence of hidden terminals. As a preliminary step, each wireless station computes the schedule length after gathering information about the number of flows in its neighbourhood. Then, a combination of deterministic and random backoffs are used to reach a collision-free schedule. A deterministic backoff is used after successful transmissions and a random backoff is used otherwise. It is explained that the short acknowledgement control packets can easily result i…

指数退避：元稳定性与隐式准入控制

Exponential Backoff: Meta-Stability and Implicit Admission Control

本文分析指数退避算法，证明其具有元稳定性，导致系统在长时间尺度上振荡且仅部分源访问信道，形成隐式准入控制，并提供理论工具与可复现的数值实验。