This
thesis addresses the study of a network time synchronization (NTS) algorithm
for wireless Ad Hoc networks (WAHNs), and the design and analysis of a medium
access control (MAC) paradigm referred to as code-based scheduling. We present
a novel network time-synchronization algorithm referred to as Clocksampling Mutual Network Synchronization (CSMNS) that
shows excellent scalability, accuracy and low implementation complexity. CSMNS is evaluated analytically and numerically in single-hop and
multi-hop scenarios. The performance of CSMNS is also
compared to the performance of the timing synchronization function (TSF)
of the IEEE 802.11 standard of which, to the best of our knowledge we are the
first to provide performance results in a multi-hop scenario. CSMNS has the
ability to make the time difference among the clocks in the network converge to
a common and small value by utilizing the timing information carried on beacons
originated in any point of the network. CSMSN also reduces the need for
constant refreshment of the timing information.
General principles of code-based scheduling are investigated and particular examples A MAC strategy we
call code-based scheduling is proposed that utilizes the codewords
of codes traditionally used for channel coding purposes. We use a novel
approach that utilizes coding theory concepts to devise a scheduling strategy that
allows for the possibility to guarantee a minimum level of performance in the
nodes of a WAHN.based on Reed-Solomon and Hermitian
codes are evaluated analytically and numerically in terms of their average,
minimum and maximum delay and throughput performance. Additionally, a metric
that can be used to identify better codes for the purpose of
codebased scheduling is identified.
The average performance of a large family of code-based scheduling protocols is
analytically compared to the one obtained by slotted-ALOHA. A code-selection
algorithm is proposed that can improve the average throughput of code-based
scheduling when the number of nodes in the network is larger than the number of
code-words available. Finally
a hybrid code-based contention-based scheduling protocol is discussed and
evaluated, which combines the performance guarantee advantage of a code-based
scheduling approach with the better average performance of contention-based
scheduling protocols with feedback.