We consider a synchronous broadcast communication channel. This channel is used by a large population of users abiding to a random-access protocol of the Capetanakis-Tsybakov-Mikhailov type, and at the same time by a cheater, with distinguished characteristics and behavior. We assume that the cheater ignores the protocol used by the other users and model his transmission attempts as a Bernoulli process. We then obtain throughput and mean delay results for both the original protocol-abiding population and the cheater. The results show that the cheater's delay is considerably lower than that of the protocol abiding population. However, there is no abrupt failure of the protocol. The stability region of the original protocol is diminished, but the decrease is almost linear with the transmission rate of the cheater. Notice that this model and analysis can serve in studying and suggesting non-obvious solutions to random-access environments with heavy users (e.g., gateways), station malfunctions, and deliberate user behavior. The obtained difference in performance can also be exploited in order to provide various degrees of priority to specific users.
[1]
J. Capetanakis,et al.
Generalized TDMA: The Multi-Accessing Tree Protocol
,
1979,
IEEE Trans. Commun..
[2]
George C. Polyzos,et al.
A queueing theoretic methodology for the analysis of separable conflict resolution algorithms with variable length elementary events
,
1994,
Queueing Syst. Theory Appl..
[3]
Dimitri P. Bertsekas,et al.
Data Networks
,
1986
.
[4]
George C. Polyzos,et al.
A queueing theoretic approach to the delay analysis for the FCFS 0.487 conflict resolution algorithm
,
1993,
IEEE Trans. Inf. Theory.
[5]
George C. Polyzos,et al.
Performance Analysis of Finite Nonhomogeneous Population Tree Conflict Resolution Algorithms Using Constant Size Window Access
,
1987,
IEEE Trans. Commun..
[6]
Moshe Sidi,et al.
Discrete-Time Priority Queues with Partial Interference
,
1987,
IEEE J. Sel. Areas Commun..