Datacycle is an information delivery system designed in Bellcore . It uses broadcast delivery and multiple channels of an optical fiber to provide a large volume of information to many users. In this system, the mean access time depends on the broadcast delivery schedule because different information items (called pages) may have different popularity and there are multiple channels for concurrent broadcast delivery. In this paper, we design broadcast delivery schedules for M channels where M 2 and our objective is to minimize the mean access time. We show that this design problem can be divided into two subproblems: (1) divide the pages into M partitions such that the pages of each partition are broadcast in a distinct channel and (2) determine a broadcast schedule for the pages of each partition. We analyze and solve these subproblems, and we demonstrate that the schedules found can nearly reach a lower bound on the minimal mean access time.

In uplink data communication in wireless networks, a portable computer may retransmit a packet multiple times before the base station receives the correct one. Each retransmission consumes communication bandwidth and battery energy of the portable computer. Therefore, it is desirable to reduce the number of retransmissions. In this paper, we propose the aggressive packet combining scheme for this purpose. The base station executes this scheme to combine multiple erroneous copies as follows: (1) perform bit-by-bit majority voting on the erroneous copies to produce a combined packet, (2) identify the least reliable bits in this combined packet, and (3) search the correct bit pattern for these bits. Then the base station may recover the correct packet, thereby reducing the mean number of retransmissions. The proposed scheme has several advantages: (1) it is more powerful than the majority packet combining scheme, (2) it can complement many existing ARQ protocols to improve their performance, (3) it does not add additional bits to the packet and hence it does not consume extra bandwidth in the wireless channel, and (4) it is only executed by the base station and hence the portable transceiver can be kept simple. The simulation results show that the proposed scheme is more bandwidth-efficient and energy-efficient than the majority packet combining scheme.

Packet concentrators are used in many high-speed computer communication systems such as fast packet switches. In these systems, the time available for concentration is very short. It is therefore desirable to realize the packet concentrators as hardware chips for fast concentration. The knockout concentrator was proposed for hardware realization. In this paper, we improve this concentrator to reduce the probability of packet loss, and the improved concentrator is called wraparound knockout concentrator. This concentrator has several wraparound paths within it, and it does not require any additional pin per chip. After contention among the packets in a slot, each winner goes to a distinct output, some losers circulate along the wraparound paths for contention in the subsequent slot, and the remaining losers are discarded. In this manner, some losers are not discarded immediately and they still have the chance to go to the outputs in the subsequent slot, thereby reducing the probability of packet loss. We analyze the number of logic gates required and the probability of packet loss. The numerical results show that if the proposed concentrator has a few wraparound paths, the probability of packet loss can already be reduced by orders of magnitude.

The quality of a modular software system depends on the reliabilities of the software modules and the software operational profile. When the software is designed for multiple customers having different operational profiles, different customers may experience different software quality. It is important to ensure that no customer will suffer from a poor software quality. In this paper, we formulate and solve three reliability allocation problems for modular software system designed for multiple customers. In these reliability allocation problems, we consider the software operational profile of every customer while fulfilling such practical constraints as cost budget and software quality requirement. The numerical results show that when the operational profiles of the customers are more skewed, it is more beneficial to take their operational profiles into account.