The need for broadcast encryption arises when a sender wishes to securely distribute messages to varying subsets of receivers, using a broadcast channel, for instance in a pay-TV scenario. This is done by selecting subsets of users and giving all users in the same subset a common decryption key. The subsets will in general be overlapping so that each user belongs to many subsets and has several different decryption keys. When the sender wants to send a message to some users, the message is encrypted using keys that those users have. In this thesis we describe some broadcast encryption schemes that have been proposed in the literature. We focus on stateless schemes which do not require receivers to update their decryption keys after the initial keys have been received; particularly we concentrate on the Subset Difference (SD) scheme. We consider the effects that the logical placement of the receivers in the tree structure used by the SD scheme has on the number of required transmissions for each message. Bounds for the number of required transmissions are derived based on the adjacency of receivers in the tree structure. The tree structure itself is also studied, also resulting in bounds on the number of required transmissions based on the placement of the users in the tree structure. By allowing a slight discrepancy between the set of receivers that the sender intends to send to and the set of receivers that actually can decrypt the message, we can reduce the cost in number of transmissions per message. We use the concept of distortion to quantify the discrepancy and develop three simple algorithms to illustrate how the cost and distortion are related. / <p>Report code: LIU-Tek-Lic-2005:70.</p>

No description available.

No description available.

No description available.

No description available.

No description available.

Vehicles are more frequently being built with hardware that supports wireless communica- tion capabilities. Dedicated short-range communications (DSRC) is a standard that enables the hardware on vehicles to communicate with one another directly rather than through external infrastructure such as a cellular tower. With DSRC supporting small-range communications, multi-hop routing is utilized when a packet needs to reach a long-range destination. A vehicular ad-hoc network (VANET) broadcast protocol (VBP) was developed. This thesis introduces VBP, an open-source framework for simulating multi-hop routing on mobile and wireless vehicular networks. VBP is built for the routing layer of the network simulation tool called network simulator 3 (ns-3) and contains a custom protocol that adapts to various traffic conditions on a roadway. To test VBP we ran six simulations across three traffic levels. Results confirm that VBP successfully routes packets or queues packets when a first or next hop is not available. The development process of VBP is documented to help researchers who are trying to create a custom routing protocol for ns-3.

<p>In this thesis, a practical superposition coding scheme based on multilevel low-density parity-check (LDPC) codes is proposed for discrete memoryless broadcast channels. The simulation results show that the performance of the proposed scheme approaches the information-theoretic limits. We also propose a method for optimizing the degree distribution of multilevel LDPC codes based on the analysis of EXIT functions.</p> / Master of Applied Science (MASc)

No description available.

A form of data transmission which is increasing in popularity is satellite communication. In order to insure that data is received correctly, certain error control strategies are employed. In packetized transmission, automatic-repeat-request (ARQ) schemes and error correcting codes have been employed. Currently a combination of both is not employed by the AX.25 protocol. The purpose of this project is to determine which error correcting code should be combined with ARQ to provide the optimum performance. This project investigates the performance of the (7,4) Hamming code, (23,12) Golay code, Reed-Solomon, RS , codes over the Galois fields of 16, GF(16), and 256, GF(256), elements, and common rate convolutional codes of various constraint lengths. The codes are evaluated on three primary criteria, which include: throughput vs. input bit error rate, coding complexity, and burst error performance. The class of Reed-Solomon codes over GF(256) was chosen to be superior due to the throughput and burst error performance. The encoding complexity is small, but the decoding is more complicated. The conclusion is that error correcting codes should be employed in ARQ satellite systems. However, the error correcting strength of the code must be determined by observing the channel characteristics. / Master of Science